Add files using upload-large-folder tool
Browse filesThis view is limited to 50 files because it contains too many changes. See raw diff
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d180/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d20/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d20/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d20/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d200/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d200/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d200/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d220/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d220/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d220/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d240/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d240/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d240/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d260/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d260/metrics.jsonl +91 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d260/runner.log +91 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d40/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d40/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d40/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d60/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d60/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d60/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d80/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d80/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d80/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d100/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d100/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d100/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d120/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d120/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d120/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d140/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d140/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d140/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d160/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d160/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d160/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d180/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d180/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d180/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d20/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d20/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d20/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200/fid_retry.json +75 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200/fid_retry.log +3 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200/metrics.jsonl +0 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200/runner.log +502 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d220/config.json +25 -0
- results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d220/metrics.jsonl +0 -0
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d180/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 83s | train_loss=1.0019 | fid=285.7530
|
| 3 |
+
step 10000 | 186s | train_loss=0.7616
|
| 4 |
+
step 20000 | 295s | train_loss=0.6333
|
| 5 |
+
step 30000 | 404s | train_loss=0.5304
|
| 6 |
+
step 40000 | 512s | train_loss=0.4425
|
| 7 |
+
step 50000 | 621s | train_loss=0.3667
|
| 8 |
+
step 60000 | 730s | train_loss=0.2998
|
| 9 |
+
step 70000 | 838s | train_loss=0.2454
|
| 10 |
+
step 80000 | 947s | train_loss=0.2078
|
| 11 |
+
step 90000 | 1055s | train_loss=0.1897
|
| 12 |
+
step 100000 | 1243s | train_loss=0.1572 | fid=195.1559
|
| 13 |
+
step 110000 | 1346s | train_loss=0.1464
|
| 14 |
+
step 120000 | 1454s | train_loss=0.1342
|
| 15 |
+
step 130000 | 1561s | train_loss=0.1335
|
| 16 |
+
step 140000 | 1670s | train_loss=0.1192
|
| 17 |
+
step 150000 | 1778s | train_loss=0.1105
|
| 18 |
+
step 160000 | 1887s | train_loss=0.1097
|
| 19 |
+
step 170000 | 1995s | train_loss=0.1053
|
| 20 |
+
step 180000 | 2103s | train_loss=0.1055
|
| 21 |
+
step 190000 | 2211s | train_loss=0.0956
|
| 22 |
+
step 200000 | 2400s | train_loss=0.0958 | fid=236.6682
|
| 23 |
+
step 210000 | 2506s | train_loss=0.0928
|
| 24 |
+
step 220000 | 2610s | train_loss=0.0914
|
| 25 |
+
step 230000 | 2718s | train_loss=0.0850
|
| 26 |
+
step 240000 | 2827s | train_loss=0.0827
|
| 27 |
+
step 250000 | 2935s | train_loss=0.0833
|
| 28 |
+
step 260000 | 3044s | train_loss=0.0828
|
| 29 |
+
step 270000 | 3152s | train_loss=0.0793
|
| 30 |
+
step 280000 | 3261s | train_loss=0.0795
|
| 31 |
+
step 290000 | 3370s | train_loss=0.0758
|
| 32 |
+
step 300000 | 3558s | train_loss=0.0752 | fid=245.7375
|
| 33 |
+
step 310000 | 3664s | train_loss=0.0725
|
| 34 |
+
step 320000 | 3773s | train_loss=0.0701
|
| 35 |
+
step 330000 | 3877s | train_loss=0.0722
|
| 36 |
+
step 340000 | 3985s | train_loss=0.0702
|
| 37 |
+
step 350000 | 4094s | train_loss=0.0659
|
| 38 |
+
step 360000 | 4202s | train_loss=0.0664
|
| 39 |
+
step 370000 | 4311s | train_loss=0.0668
|
| 40 |
+
step 380000 | 4419s | train_loss=0.0632
|
| 41 |
+
step 390000 | 4528s | train_loss=0.0643
|
| 42 |
+
step 400000 | 4715s | train_loss=0.0609 | fid=280.9470
|
| 43 |
+
step 410000 | 4821s | train_loss=0.0588
|
| 44 |
+
step 420000 | 4930s | train_loss=0.0575
|
| 45 |
+
step 430000 | 5039s | train_loss=0.0570
|
| 46 |
+
step 440000 | 5142s | train_loss=0.0568
|
| 47 |
+
step 450000 | 5251s | train_loss=0.0571
|
| 48 |
+
step 460000 | 5359s | train_loss=0.0564
|
| 49 |
+
step 470000 | 5468s | train_loss=0.0562
|
| 50 |
+
step 480000 | 5577s | train_loss=0.0560
|
| 51 |
+
step 490000 | 5685s | train_loss=0.0599
|
| 52 |
+
step 500000 | 5873s | train_loss=0.0582 | fid=304.7993
|
| 53 |
+
step 510000 | 5981s | train_loss=0.0541
|
| 54 |
+
step 520000 | 6087s | train_loss=0.0530
|
| 55 |
+
step 530000 | 6196s | train_loss=0.0545
|
| 56 |
+
step 540000 | 6304s | train_loss=0.0529
|
| 57 |
+
step 550000 | 6408s | train_loss=0.0533
|
| 58 |
+
step 560000 | 6516s | train_loss=0.0500
|
| 59 |
+
step 570000 | 6625s | train_loss=0.0495
|
| 60 |
+
step 580000 | 6733s | train_loss=0.0526
|
| 61 |
+
step 590000 | 6841s | train_loss=0.0497
|
| 62 |
+
step 600000 | 7029s | train_loss=0.0506 | fid=285.8958
|
| 63 |
+
step 610000 | 7138s | train_loss=0.0475
|
| 64 |
+
step 620000 | 7244s | train_loss=0.0465
|
| 65 |
+
step 630000 | 7353s | train_loss=0.0473
|
| 66 |
+
step 640000 | 7461s | train_loss=0.0466
|
| 67 |
+
step 650000 | 7571s | train_loss=0.0511
|
| 68 |
+
step 660000 | 7674s | train_loss=0.0483
|
| 69 |
+
step 670000 | 7783s | train_loss=0.0491
|
| 70 |
+
step 680000 | 7891s | train_loss=0.0448
|
| 71 |
+
step 690000 | 8000s | train_loss=0.0446
|
| 72 |
+
step 700000 | 8188s | train_loss=0.0461 | fid=273.9181
|
| 73 |
+
step 710000 | 8296s | train_loss=0.0441
|
| 74 |
+
step 720000 | 8403s | train_loss=0.0434
|
| 75 |
+
step 730000 | 8511s | train_loss=0.0443
|
| 76 |
+
step 740000 | 8619s | train_loss=0.0443
|
| 77 |
+
step 750000 | 8728s | train_loss=0.0415
|
| 78 |
+
step 760000 | 8837s | train_loss=0.0420
|
| 79 |
+
step 770000 | 8940s | train_loss=0.0409
|
| 80 |
+
step 780000 | 9049s | train_loss=0.0465
|
| 81 |
+
step 790000 | 9157s | train_loss=0.0426
|
| 82 |
+
step 800000 | 9345s | train_loss=0.0405 | fid=271.5777
|
| 83 |
+
step 810000 | 9454s | train_loss=0.0418
|
| 84 |
+
step 820000 | 9560s | train_loss=0.0401
|
| 85 |
+
step 830000 | 9668s | train_loss=0.0418
|
| 86 |
+
step 840000 | 9777s | train_loss=0.0395
|
| 87 |
+
step 850000 | 9886s | train_loss=0.0426
|
| 88 |
+
step 860000 | 9995s | train_loss=0.0422
|
| 89 |
+
step 870000 | 10102s | train_loss=0.0406
|
| 90 |
+
step 880000 | 10208s | train_loss=0.0393
|
| 91 |
+
step 890000 | 10315s | train_loss=0.0404
|
| 92 |
+
step 900000 | 10505s | train_loss=0.0379 | fid=259.0478
|
| 93 |
+
step 910000 | 10613s | train_loss=0.0404
|
| 94 |
+
step 920000 | 10719s | train_loss=0.0372
|
| 95 |
+
step 930000 | 10828s | train_loss=0.0388
|
| 96 |
+
step 940000 | 10936s | train_loss=0.0392
|
| 97 |
+
step 950000 | 11044s | train_loss=0.0365
|
| 98 |
+
step 960000 | 11153s | train_loss=0.0372
|
| 99 |
+
step 970000 | 11260s | train_loss=0.0377
|
| 100 |
+
step 980000 | 11368s | train_loss=0.0386
|
| 101 |
+
step 990000 | 11466s | train_loss=0.0368
|
| 102 |
+
step 1000000 | 11656s | train_loss=0.0359 | fid=246.0043
|
| 103 |
+
step 1010000 | 11764s | train_loss=0.0371
|
| 104 |
+
step 1020000 | 11873s | train_loss=0.0350
|
| 105 |
+
step 1030000 | 11979s | train_loss=0.0366
|
| 106 |
+
step 1040000 | 12088s | train_loss=0.0357
|
| 107 |
+
step 1050000 | 12196s | train_loss=0.0353
|
| 108 |
+
step 1060000 | 12305s | train_loss=0.0394
|
| 109 |
+
step 1070000 | 12413s | train_loss=0.0346
|
| 110 |
+
step 1080000 | 12522s | train_loss=0.0377
|
| 111 |
+
step 1090000 | 12627s | train_loss=0.0359
|
| 112 |
+
step 1100000 | 12814s | train_loss=0.0329 | fid=235.3345
|
| 113 |
+
step 1110000 | 12923s | train_loss=0.0327
|
| 114 |
+
step 1120000 | 13031s | train_loss=0.0328
|
| 115 |
+
step 1130000 | 13137s | train_loss=0.0377
|
| 116 |
+
step 1140000 | 13245s | train_loss=0.0333
|
| 117 |
+
step 1150000 | 13353s | train_loss=0.0348
|
| 118 |
+
step 1160000 | 13461s | train_loss=0.0359
|
| 119 |
+
step 1170000 | 13570s | train_loss=0.0359
|
| 120 |
+
step 1180000 | 13663s | train_loss=0.0335
|
| 121 |
+
step 1190000 | 13750s | train_loss=0.0338
|
| 122 |
+
step 1200000 | 13906s | train_loss=0.0324 | fid=228.7945
|
| 123 |
+
step 1210000 | 13984s | train_loss=0.0326
|
| 124 |
+
step 1220000 | 14074s | train_loss=0.0339
|
| 125 |
+
step 1230000 | 14154s | train_loss=0.0340
|
| 126 |
+
step 1240000 | 14243s | train_loss=0.0331
|
| 127 |
+
step 1250000 | 14333s | train_loss=0.0322
|
| 128 |
+
step 1260000 | 14423s | train_loss=0.0334
|
| 129 |
+
step 1270000 | 14513s | train_loss=0.0346
|
| 130 |
+
step 1280000 | 14603s | train_loss=0.0317
|
| 131 |
+
step 1290000 | 14690s | train_loss=0.0314
|
| 132 |
+
step 1300000 | 14848s | train_loss=0.0317 | fid=223.5888
|
| 133 |
+
step 1310000 | 14938s | train_loss=0.0313
|
| 134 |
+
step 1320000 | 15025s | train_loss=0.0345
|
| 135 |
+
step 1330000 | 15114s | train_loss=0.0344
|
| 136 |
+
step 1340000 | 15201s | train_loss=0.0325
|
| 137 |
+
step 1350000 | 15291s | train_loss=0.0358
|
| 138 |
+
step 1360000 | 15381s | train_loss=0.0311
|
| 139 |
+
step 1370000 | 15471s | train_loss=0.0302
|
| 140 |
+
step 1380000 | 15560s | train_loss=0.0310
|
| 141 |
+
step 1390000 | 15649s | train_loss=0.0319
|
| 142 |
+
step 1400000 | 15808s | train_loss=0.0321 | fid=214.1127
|
| 143 |
+
step 1410000 | 15896s | train_loss=0.0301
|
| 144 |
+
step 1420000 | 15986s | train_loss=0.0315
|
| 145 |
+
step 1430000 | 16073s | train_loss=0.0298
|
| 146 |
+
step 1440000 | 16160s | train_loss=0.0309
|
| 147 |
+
step 1450000 | 16250s | train_loss=0.0301
|
| 148 |
+
step 1460000 | 16341s | train_loss=0.0301
|
| 149 |
+
step 1470000 | 16431s | train_loss=0.0311
|
| 150 |
+
step 1480000 | 16519s | train_loss=0.0303
|
| 151 |
+
step 1490000 | 16609s | train_loss=0.0327
|
| 152 |
+
step 1500000 | 16768s | train_loss=0.0298 | fid=200.5839
|
| 153 |
+
step 1510000 | 16857s | train_loss=0.0320
|
| 154 |
+
step 1520000 | 16947s | train_loss=0.0318
|
| 155 |
+
step 1530000 | 17036s | train_loss=0.0337
|
| 156 |
+
step 1540000 | 17120s | train_loss=0.0287
|
| 157 |
+
step 1550000 | 17210s | train_loss=0.0311
|
| 158 |
+
step 1560000 | 17300s | train_loss=0.0303
|
| 159 |
+
step 1570000 | 17390s | train_loss=0.0270
|
| 160 |
+
step 1580000 | 17477s | train_loss=0.0302
|
| 161 |
+
step 1590000 | 17567s | train_loss=0.0288
|
| 162 |
+
step 1600000 | 17726s | train_loss=0.0302 | fid=196.9809
|
| 163 |
+
step 1610000 | 17816s | train_loss=0.0287
|
| 164 |
+
step 1620000 | 17905s | train_loss=0.0289
|
| 165 |
+
step 1630000 | 17995s | train_loss=0.0288
|
| 166 |
+
step 1640000 | 18082s | train_loss=0.0296
|
| 167 |
+
step 1650000 | 18169s | train_loss=0.0288
|
| 168 |
+
step 1660000 | 18259s | train_loss=0.0282
|
| 169 |
+
step 1670000 | 18349s | train_loss=0.0286
|
| 170 |
+
step 1680000 | 18438s | train_loss=0.0256
|
| 171 |
+
step 1690000 | 18528s | train_loss=0.0293
|
| 172 |
+
step 1700000 | 18686s | train_loss=0.0276 | fid=195.9055
|
| 173 |
+
step 1710000 | 18776s | train_loss=0.0266
|
| 174 |
+
step 1720000 | 18866s | train_loss=0.0272
|
| 175 |
+
step 1730000 | 18954s | train_loss=0.0259
|
| 176 |
+
step 1740000 | 19043s | train_loss=0.0262
|
| 177 |
+
step 1750000 | 19125s | train_loss=0.0286
|
| 178 |
+
step 1760000 | 19199s | train_loss=0.0269
|
| 179 |
+
step 1770000 | 19286s | train_loss=0.0272
|
| 180 |
+
step 1780000 | 19374s | train_loss=0.0288
|
| 181 |
+
step 1790000 | 19464s | train_loss=0.0274
|
| 182 |
+
step 1800000 | 19623s | train_loss=0.0294 | fid=189.8670
|
| 183 |
+
step 1810000 | 19713s | train_loss=0.0263
|
| 184 |
+
step 1820000 | 19803s | train_loss=0.0271
|
| 185 |
+
step 1830000 | 19893s | train_loss=0.0272
|
| 186 |
+
step 1840000 | 19981s | train_loss=0.0277
|
| 187 |
+
step 1850000 | 20063s | train_loss=0.0265
|
| 188 |
+
step 1860000 | 20149s | train_loss=0.0270
|
| 189 |
+
step 1870000 | 20237s | train_loss=0.0253
|
| 190 |
+
step 1880000 | 20316s | train_loss=0.0277
|
| 191 |
+
step 1890000 | 20406s | train_loss=0.0291
|
| 192 |
+
step 1900000 | 20565s | train_loss=0.0282 | fid=180.6721
|
| 193 |
+
step 1910000 | 20655s | train_loss=0.0274
|
| 194 |
+
step 1920000 | 20744s | train_loss=0.0256
|
| 195 |
+
step 1930000 | 20834s | train_loss=0.0276
|
| 196 |
+
step 1940000 | 20924s | train_loss=0.0254
|
| 197 |
+
step 1950000 | 21012s | train_loss=0.0252
|
| 198 |
+
step 1960000 | 21092s | train_loss=0.0254
|
| 199 |
+
step 1970000 | 21180s | train_loss=0.0250
|
| 200 |
+
step 1980000 | 21270s | train_loss=0.0249
|
| 201 |
+
step 1990000 | 21348s | train_loss=0.0260
|
| 202 |
+
step 2000000 | 21504s | train_loss=0.0241 | fid=181.5921
|
| 203 |
+
step 2010000 | 21594s | train_loss=0.0277
|
| 204 |
+
step 2020000 | 21684s | train_loss=0.0261
|
| 205 |
+
step 2030000 | 21774s | train_loss=0.0271
|
| 206 |
+
step 2040000 | 21864s | train_loss=0.0259
|
| 207 |
+
step 2050000 | 21954s | train_loss=0.0236
|
| 208 |
+
step 2060000 | 22040s | train_loss=0.0260
|
| 209 |
+
step 2070000 | 22119s | train_loss=0.0256
|
| 210 |
+
step 2080000 | 22210s | train_loss=0.0239
|
| 211 |
+
step 2090000 | 22300s | train_loss=0.0250
|
| 212 |
+
step 2100000 | 22444s | train_loss=0.0261 | fid=170.5414
|
| 213 |
+
step 2110000 | 22534s | train_loss=0.0249
|
| 214 |
+
step 2120000 | 22624s | train_loss=0.0247
|
| 215 |
+
step 2130000 | 22714s | train_loss=0.0264
|
| 216 |
+
step 2140000 | 22803s | train_loss=0.0256
|
| 217 |
+
step 2150000 | 22893s | train_loss=0.0249
|
| 218 |
+
step 2160000 | 22981s | train_loss=0.0242
|
| 219 |
+
step 2170000 | 23060s | train_loss=0.0259
|
| 220 |
+
step 2180000 | 23148s | train_loss=0.0246
|
| 221 |
+
step 2190000 | 23237s | train_loss=0.0238
|
| 222 |
+
step 2200000 | 23396s | train_loss=0.0234 | fid=160.8073
|
| 223 |
+
step 2210000 | 23482s | train_loss=0.0234
|
| 224 |
+
step 2220000 | 23562s | train_loss=0.0254
|
| 225 |
+
step 2230000 | 23652s | train_loss=0.0241
|
| 226 |
+
step 2240000 | 23742s | train_loss=0.0233
|
| 227 |
+
step 2250000 | 23832s | train_loss=0.0291
|
| 228 |
+
step 2260000 | 23922s | train_loss=0.0238
|
| 229 |
+
step 2270000 | 24008s | train_loss=0.0249
|
| 230 |
+
step 2280000 | 24084s | train_loss=0.0229
|
| 231 |
+
step 2290000 | 24174s | train_loss=0.0253
|
| 232 |
+
step 2300000 | 24333s | train_loss=0.0256 | fid=160.3043
|
| 233 |
+
step 2310000 | 24423s | train_loss=0.0236
|
| 234 |
+
step 2320000 | 24513s | train_loss=0.0241
|
| 235 |
+
step 2330000 | 24592s | train_loss=0.0238
|
| 236 |
+
step 2340000 | 24682s | train_loss=0.0259
|
| 237 |
+
step 2350000 | 24772s | train_loss=0.0230
|
| 238 |
+
step 2360000 | 24860s | train_loss=0.0279
|
| 239 |
+
step 2370000 | 24950s | train_loss=0.0243
|
| 240 |
+
step 2380000 | 25038s | train_loss=0.0235
|
| 241 |
+
step 2390000 | 25119s | train_loss=0.0230
|
| 242 |
+
step 2400000 | 25278s | train_loss=0.0247 | fid=151.5119
|
| 243 |
+
step 2410000 | 25368s | train_loss=0.0226
|
| 244 |
+
step 2420000 | 25458s | train_loss=0.0231
|
| 245 |
+
step 2430000 | 25548s | train_loss=0.0254
|
| 246 |
+
step 2440000 | 25632s | train_loss=0.0225
|
| 247 |
+
step 2450000 | 25718s | train_loss=0.0239
|
| 248 |
+
step 2460000 | 25806s | train_loss=0.0237
|
| 249 |
+
step 2470000 | 25896s | train_loss=0.0252
|
| 250 |
+
step 2480000 | 25987s | train_loss=0.0223
|
| 251 |
+
step 2490000 | 26068s | train_loss=0.0233
|
| 252 |
+
step 2500000 | 26220s | train_loss=0.0210 | fid=154.7370
|
| 253 |
+
step 2510000 | 26310s | train_loss=0.0219
|
| 254 |
+
step 2520000 | 26400s | train_loss=0.0240
|
| 255 |
+
step 2530000 | 26490s | train_loss=0.0230
|
| 256 |
+
step 2540000 | 26580s | train_loss=0.0224
|
| 257 |
+
step 2550000 | 26669s | train_loss=0.0234
|
| 258 |
+
step 2560000 | 26740s | train_loss=0.0216
|
| 259 |
+
step 2570000 | 26830s | train_loss=0.0219
|
| 260 |
+
step 2580000 | 26920s | train_loss=0.0229
|
| 261 |
+
step 2590000 | 27010s | train_loss=0.0223
|
| 262 |
+
step 2600000 | 27156s | train_loss=0.0222 | fid=146.7394
|
| 263 |
+
step 2610000 | 27246s | train_loss=0.0232
|
| 264 |
+
step 2620000 | 27336s | train_loss=0.0208
|
| 265 |
+
step 2630000 | 27426s | train_loss=0.0222
|
| 266 |
+
step 2640000 | 27516s | train_loss=0.0236
|
| 267 |
+
step 2650000 | 27606s | train_loss=0.0231
|
| 268 |
+
step 2660000 | 27694s | train_loss=0.0222
|
| 269 |
+
step 2670000 | 27777s | train_loss=0.0224
|
| 270 |
+
step 2680000 | 27857s | train_loss=0.0220
|
| 271 |
+
step 2690000 | 27947s | train_loss=0.0218
|
| 272 |
+
step 2700000 | 28095s | train_loss=0.0231 | fid=153.1720
|
| 273 |
+
step 2710000 | 28184s | train_loss=0.0242
|
| 274 |
+
step 2720000 | 28274s | train_loss=0.0226
|
| 275 |
+
step 2730000 | 28363s | train_loss=0.0200
|
| 276 |
+
step 2740000 | 28453s | train_loss=0.0240
|
| 277 |
+
step 2750000 | 28541s | train_loss=0.0201
|
| 278 |
+
step 2760000 | 28630s | train_loss=0.0201
|
| 279 |
+
step 2770000 | 28720s | train_loss=0.0207
|
| 280 |
+
step 2780000 | 28810s | train_loss=0.0200
|
| 281 |
+
step 2790000 | 28889s | train_loss=0.0206
|
| 282 |
+
step 2800000 | 29047s | train_loss=0.0214 | fid=145.0943
|
| 283 |
+
step 2810000 | 29124s | train_loss=0.0219
|
| 284 |
+
step 2820000 | 29214s | train_loss=0.0214
|
| 285 |
+
step 2830000 | 29304s | train_loss=0.0217
|
| 286 |
+
step 2840000 | 29394s | train_loss=0.0218
|
| 287 |
+
step 2850000 | 29482s | train_loss=0.0204
|
| 288 |
+
step 2860000 | 29572s | train_loss=0.0203
|
| 289 |
+
step 2870000 | 29662s | train_loss=0.0232
|
| 290 |
+
step 2880000 | 29752s | train_loss=0.0214
|
| 291 |
+
step 2890000 | 29842s | train_loss=0.0217
|
| 292 |
+
step 2900000 | 29984s | train_loss=0.0212 | fid=139.4337
|
| 293 |
+
step 2910000 | 30071s | train_loss=0.0212
|
| 294 |
+
step 2920000 | 30146s | train_loss=0.0214
|
| 295 |
+
step 2930000 | 30236s | train_loss=0.0204
|
| 296 |
+
step 2940000 | 30325s | train_loss=0.0214
|
| 297 |
+
step 2950000 | 30413s | train_loss=0.0221
|
| 298 |
+
step 2960000 | 30503s | train_loss=0.0198
|
| 299 |
+
step 2970000 | 30593s | train_loss=0.0195
|
| 300 |
+
step 2980000 | 30683s | train_loss=0.0206
|
| 301 |
+
step 2990000 | 30773s | train_loss=0.0215
|
| 302 |
+
step 3000000 | 30932s | train_loss=0.0215 | fid=148.1522
|
| 303 |
+
step 3010000 | 31015s | train_loss=0.0211
|
| 304 |
+
step 3020000 | 31097s | train_loss=0.0198
|
| 305 |
+
step 3030000 | 31178s | train_loss=0.0238
|
| 306 |
+
step 3040000 | 31268s | train_loss=0.0218
|
| 307 |
+
step 3050000 | 31356s | train_loss=0.0213
|
| 308 |
+
step 3060000 | 31446s | train_loss=0.0213
|
| 309 |
+
step 3070000 | 31536s | train_loss=0.0205
|
| 310 |
+
step 3080000 | 31626s | train_loss=0.0218
|
| 311 |
+
step 3090000 | 31716s | train_loss=0.0211
|
| 312 |
+
step 3100000 | 31875s | train_loss=0.0219 | fid=141.1102
|
| 313 |
+
step 3110000 | 31965s | train_loss=0.0191
|
| 314 |
+
step 3120000 | 32055s | train_loss=0.0212
|
| 315 |
+
step 3130000 | 32126s | train_loss=0.0207
|
| 316 |
+
step 3140000 | 32210s | train_loss=0.0200
|
| 317 |
+
step 3150000 | 32298s | train_loss=0.0199
|
| 318 |
+
step 3160000 | 32388s | train_loss=0.0212
|
| 319 |
+
step 3170000 | 32477s | train_loss=0.0212
|
| 320 |
+
step 3180000 | 32567s | train_loss=0.0205
|
| 321 |
+
step 3190000 | 32657s | train_loss=0.0205
|
| 322 |
+
step 3200000 | 32816s | train_loss=0.0195 | fid=133.7140
|
| 323 |
+
step 3210000 | 32906s | train_loss=0.0217
|
| 324 |
+
step 3220000 | 32996s | train_loss=0.0195
|
| 325 |
+
step 3230000 | 33087s | train_loss=0.0193
|
| 326 |
+
step 3240000 | 33156s | train_loss=0.0199
|
| 327 |
+
step 3250000 | 33246s | train_loss=0.0215
|
| 328 |
+
step 3260000 | 33336s | train_loss=0.0201
|
| 329 |
+
step 3270000 | 33426s | train_loss=0.0237
|
| 330 |
+
step 3280000 | 33516s | train_loss=0.0218
|
| 331 |
+
step 3290000 | 33606s | train_loss=0.0194
|
| 332 |
+
step 3300000 | 33764s | train_loss=0.0219 | fid=129.6199
|
| 333 |
+
step 3310000 | 33854s | train_loss=0.0199
|
| 334 |
+
step 3320000 | 33944s | train_loss=0.0197
|
| 335 |
+
step 3330000 | 34034s | train_loss=0.0193
|
| 336 |
+
step 3340000 | 34116s | train_loss=0.0206
|
| 337 |
+
step 3350000 | 34191s | train_loss=0.0207
|
| 338 |
+
step 3360000 | 34277s | train_loss=0.0190
|
| 339 |
+
step 3370000 | 34366s | train_loss=0.0214
|
| 340 |
+
step 3380000 | 34456s | train_loss=0.0196
|
| 341 |
+
step 3390000 | 34546s | train_loss=0.0196
|
| 342 |
+
step 3400000 | 34705s | train_loss=0.0192 | fid=137.1433
|
| 343 |
+
step 3410000 | 34795s | train_loss=0.0210
|
| 344 |
+
step 3420000 | 34885s | train_loss=0.0198
|
| 345 |
+
step 3430000 | 34975s | train_loss=0.0200
|
| 346 |
+
step 3440000 | 35063s | train_loss=0.0195
|
| 347 |
+
step 3450000 | 35140s | train_loss=0.0192
|
| 348 |
+
step 3460000 | 35227s | train_loss=0.0187
|
| 349 |
+
step 3470000 | 35306s | train_loss=0.0196
|
| 350 |
+
step 3480000 | 35396s | train_loss=0.0185
|
| 351 |
+
step 3490000 | 35486s | train_loss=0.0200
|
| 352 |
+
step 3500000 | 35645s | train_loss=0.0189 | fid=130.2748
|
| 353 |
+
step 3510000 | 35735s | train_loss=0.0189
|
| 354 |
+
step 3520000 | 35825s | train_loss=0.0191
|
| 355 |
+
step 3530000 | 35914s | train_loss=0.0196
|
| 356 |
+
step 3540000 | 36002s | train_loss=0.0178
|
| 357 |
+
step 3550000 | 36092s | train_loss=0.0198
|
| 358 |
+
step 3560000 | 36170s | train_loss=0.0182
|
| 359 |
+
step 3570000 | 36260s | train_loss=0.0203
|
| 360 |
+
step 3580000 | 36341s | train_loss=0.0200
|
| 361 |
+
step 3590000 | 36431s | train_loss=0.0202
|
| 362 |
+
step 3600000 | 36589s | train_loss=0.0193 | fid=133.5527
|
| 363 |
+
step 3610000 | 36679s | train_loss=0.0195
|
| 364 |
+
step 3620000 | 36769s | train_loss=0.0180
|
| 365 |
+
step 3630000 | 36857s | train_loss=0.0200
|
| 366 |
+
step 3640000 | 36946s | train_loss=0.0198
|
| 367 |
+
step 3650000 | 37037s | train_loss=0.0209
|
| 368 |
+
step 3660000 | 37125s | train_loss=0.0192
|
| 369 |
+
step 3670000 | 37204s | train_loss=0.0200
|
| 370 |
+
step 3680000 | 37294s | train_loss=0.0192
|
| 371 |
+
step 3690000 | 37377s | train_loss=0.0185
|
| 372 |
+
step 3700000 | 37528s | train_loss=0.0179 | fid=133.0020
|
| 373 |
+
step 3710000 | 37618s | train_loss=0.0186
|
| 374 |
+
step 3720000 | 37708s | train_loss=0.0215
|
| 375 |
+
step 3730000 | 37796s | train_loss=0.0186
|
| 376 |
+
step 3740000 | 37886s | train_loss=0.0184
|
| 377 |
+
step 3750000 | 37975s | train_loss=0.0201
|
| 378 |
+
step 3760000 | 38065s | train_loss=0.0188
|
| 379 |
+
step 3770000 | 38148s | train_loss=0.0208
|
| 380 |
+
step 3780000 | 38227s | train_loss=0.0197
|
| 381 |
+
step 3790000 | 38317s | train_loss=0.0183
|
| 382 |
+
step 3800000 | 38474s | train_loss=0.0185 | fid=128.3379
|
| 383 |
+
step 3810000 | 38564s | train_loss=0.0203
|
| 384 |
+
step 3820000 | 38653s | train_loss=0.0181
|
| 385 |
+
step 3830000 | 38742s | train_loss=0.0185
|
| 386 |
+
step 3840000 | 38832s | train_loss=0.0185
|
| 387 |
+
step 3850000 | 38922s | train_loss=0.0190
|
| 388 |
+
step 3860000 | 39012s | train_loss=0.0198
|
| 389 |
+
step 3870000 | 39102s | train_loss=0.0205
|
| 390 |
+
step 3880000 | 39188s | train_loss=0.0191
|
| 391 |
+
step 3890000 | 39277s | train_loss=0.0191
|
| 392 |
+
step 3900000 | 39436s | train_loss=0.0179 | fid=140.1177
|
| 393 |
+
step 3910000 | 39524s | train_loss=0.0200
|
| 394 |
+
step 3920000 | 39614s | train_loss=0.0192
|
| 395 |
+
step 3930000 | 39702s | train_loss=0.0184
|
| 396 |
+
step 3940000 | 39792s | train_loss=0.0192
|
| 397 |
+
step 3950000 | 39882s | train_loss=0.0192
|
| 398 |
+
step 3960000 | 39972s | train_loss=0.0185
|
| 399 |
+
step 3970000 | 40061s | train_loss=0.0198
|
| 400 |
+
step 3980000 | 40148s | train_loss=0.0176
|
| 401 |
+
step 3990000 | 40236s | train_loss=0.0208
|
| 402 |
+
step 4000000 | 40395s | train_loss=0.0167 | fid=131.6967
|
| 403 |
+
step 4010000 | 40486s | train_loss=0.0186
|
| 404 |
+
step 4020000 | 40561s | train_loss=0.0192
|
| 405 |
+
step 4030000 | 40649s | train_loss=0.0192
|
| 406 |
+
step 4040000 | 40739s | train_loss=0.0192
|
| 407 |
+
step 4050000 | 40829s | train_loss=0.0178
|
| 408 |
+
step 4060000 | 40919s | train_loss=0.0177
|
| 409 |
+
step 4070000 | 41009s | train_loss=0.0168
|
| 410 |
+
step 4080000 | 41098s | train_loss=0.0187
|
| 411 |
+
step 4090000 | 41175s | train_loss=0.0180
|
| 412 |
+
step 4100000 | 41333s | train_loss=0.0184 | fid=136.6324
|
| 413 |
+
step 4110000 | 41423s | train_loss=0.0169
|
| 414 |
+
step 4120000 | 41510s | train_loss=0.0203
|
| 415 |
+
step 4130000 | 41595s | train_loss=0.0179
|
| 416 |
+
step 4140000 | 41676s | train_loss=0.0196
|
| 417 |
+
step 4150000 | 41766s | train_loss=0.0170
|
| 418 |
+
step 4160000 | 41855s | train_loss=0.0174
|
| 419 |
+
step 4170000 | 41945s | train_loss=0.0175
|
| 420 |
+
step 4180000 | 42035s | train_loss=0.0210
|
| 421 |
+
step 4190000 | 42122s | train_loss=0.0193
|
| 422 |
+
step 4200000 | 42266s | train_loss=0.0173 | fid=123.7924
|
| 423 |
+
step 4210000 | 42356s | train_loss=0.0175
|
| 424 |
+
step 4220000 | 42444s | train_loss=0.0197
|
| 425 |
+
step 4230000 | 42534s | train_loss=0.0176
|
| 426 |
+
step 4240000 | 42624s | train_loss=0.0178
|
| 427 |
+
step 4250000 | 42700s | train_loss=0.0184
|
| 428 |
+
step 4260000 | 42785s | train_loss=0.0174
|
| 429 |
+
step 4270000 | 42875s | train_loss=0.0170
|
| 430 |
+
step 4280000 | 42965s | train_loss=0.0177
|
| 431 |
+
step 4290000 | 43055s | train_loss=0.0187
|
| 432 |
+
step 4300000 | 43199s | train_loss=0.0168 | fid=119.9332
|
| 433 |
+
step 4310000 | 43287s | train_loss=0.0169
|
| 434 |
+
step 4320000 | 43375s | train_loss=0.0168
|
| 435 |
+
step 4330000 | 43465s | train_loss=0.0179
|
| 436 |
+
step 4340000 | 43555s | train_loss=0.0200
|
| 437 |
+
step 4350000 | 43644s | train_loss=0.0186
|
| 438 |
+
step 4360000 | 43735s | train_loss=0.0174
|
| 439 |
+
step 4370000 | 43812s | train_loss=0.0195
|
| 440 |
+
step 4380000 | 43901s | train_loss=0.0187
|
| 441 |
+
step 4390000 | 43991s | train_loss=0.0190
|
| 442 |
+
step 4400000 | 44150s | train_loss=0.0179 | fid=120.3997
|
| 443 |
+
step 4410000 | 44226s | train_loss=0.0189
|
| 444 |
+
step 4420000 | 44312s | train_loss=0.0178
|
| 445 |
+
step 4430000 | 44402s | train_loss=0.0182
|
| 446 |
+
step 4440000 | 44492s | train_loss=0.0171
|
| 447 |
+
step 4450000 | 44581s | train_loss=0.0183
|
| 448 |
+
step 4460000 | 44672s | train_loss=0.0192
|
| 449 |
+
step 4470000 | 44762s | train_loss=0.0171
|
| 450 |
+
step 4480000 | 44843s | train_loss=0.0164
|
| 451 |
+
step 4490000 | 44929s | train_loss=0.0194
|
| 452 |
+
step 4500000 | 45087s | train_loss=0.0182 | fid=124.4724
|
| 453 |
+
step 4510000 | 45174s | train_loss=0.0181
|
| 454 |
+
step 4520000 | 45250s | train_loss=0.0184
|
| 455 |
+
step 4530000 | 45338s | train_loss=0.0173
|
| 456 |
+
step 4540000 | 45428s | train_loss=0.0173
|
| 457 |
+
step 4550000 | 45518s | train_loss=0.0205
|
| 458 |
+
step 4560000 | 45607s | train_loss=0.0167
|
| 459 |
+
step 4570000 | 45697s | train_loss=0.0163
|
| 460 |
+
step 4580000 | 45787s | train_loss=0.0180
|
| 461 |
+
step 4590000 | 45877s | train_loss=0.0190
|
| 462 |
+
step 4600000 | 46016s | train_loss=0.0191 | fid=117.6109
|
| 463 |
+
step 4610000 | 46104s | train_loss=0.0169
|
| 464 |
+
step 4620000 | 46190s | train_loss=0.0188
|
| 465 |
+
step 4630000 | 46268s | train_loss=0.0181
|
| 466 |
+
step 4640000 | 46358s | train_loss=0.0178
|
| 467 |
+
step 4650000 | 46448s | train_loss=0.0158
|
| 468 |
+
step 4660000 | 46538s | train_loss=0.0178
|
| 469 |
+
step 4670000 | 46627s | train_loss=0.0174
|
| 470 |
+
step 4680000 | 46717s | train_loss=0.0171
|
| 471 |
+
step 4690000 | 46807s | train_loss=0.0176
|
| 472 |
+
step 4700000 | 46967s | train_loss=0.0178 | fid=118.1452
|
| 473 |
+
step 4710000 | 47039s | train_loss=0.0169
|
| 474 |
+
step 4720000 | 47129s | train_loss=0.0161
|
| 475 |
+
step 4730000 | 47212s | train_loss=0.0165
|
| 476 |
+
step 4740000 | 47294s | train_loss=0.0195
|
| 477 |
+
step 4750000 | 47384s | train_loss=0.0175
|
| 478 |
+
step 4760000 | 47474s | train_loss=0.0169
|
| 479 |
+
step 4770000 | 47564s | train_loss=0.0182
|
| 480 |
+
step 4780000 | 47654s | train_loss=0.0177
|
| 481 |
+
step 4790000 | 47743s | train_loss=0.0180
|
| 482 |
+
step 4800000 | 47906s | train_loss=0.0161 | fid=116.7949
|
| 483 |
+
step 4810000 | 47995s | train_loss=0.0167
|
| 484 |
+
step 4820000 | 48082s | train_loss=0.0166
|
| 485 |
+
step 4830000 | 48172s | train_loss=0.0174
|
| 486 |
+
step 4840000 | 48259s | train_loss=0.0170
|
| 487 |
+
step 4850000 | 48347s | train_loss=0.0161
|
| 488 |
+
step 4860000 | 48437s | train_loss=0.0177
|
| 489 |
+
step 4870000 | 48528s | train_loss=0.0155
|
| 490 |
+
step 4880000 | 48618s | train_loss=0.0178
|
| 491 |
+
step 4890000 | 48708s | train_loss=0.0163
|
| 492 |
+
step 4900000 | 48870s | train_loss=0.0170 | fid=123.4492
|
| 493 |
+
step 4910000 | 48944s | train_loss=0.0162
|
| 494 |
+
step 4920000 | 49016s | train_loss=0.0166
|
| 495 |
+
step 4930000 | 49086s | train_loss=0.0168
|
| 496 |
+
step 4940000 | 49154s | train_loss=0.0162
|
| 497 |
+
step 4950000 | 49226s | train_loss=0.0183
|
| 498 |
+
step 4960000 | 49296s | train_loss=0.0170
|
| 499 |
+
step 4970000 | 49369s | train_loss=0.0163
|
| 500 |
+
step 4980000 | 49441s | train_loss=0.0160
|
| 501 |
+
step 4990000 | 49514s | train_loss=0.0165
|
| 502 |
+
step 5000000 | 49647s | train_loss=0.0170 | fid=115.9491
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d20/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d20/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d20",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 44,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d20/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d20/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 71s | train_loss=0.9916 | fid=182.9231
|
| 3 |
+
step 10000 | 163s | train_loss=0.9042
|
| 4 |
+
step 20000 | 258s | train_loss=0.8984
|
| 5 |
+
step 30000 | 352s | train_loss=0.8929
|
| 6 |
+
step 40000 | 447s | train_loss=0.8874
|
| 7 |
+
step 50000 | 541s | train_loss=0.8804
|
| 8 |
+
step 60000 | 636s | train_loss=0.8690
|
| 9 |
+
step 70000 | 730s | train_loss=0.8452
|
| 10 |
+
step 80000 | 824s | train_loss=0.8026
|
| 11 |
+
step 90000 | 918s | train_loss=0.7330
|
| 12 |
+
step 100000 | 1076s | train_loss=0.6286 | fid=163.6052
|
| 13 |
+
step 110000 | 1169s | train_loss=0.5100
|
| 14 |
+
step 120000 | 1262s | train_loss=0.4039
|
| 15 |
+
step 130000 | 1356s | train_loss=0.3254
|
| 16 |
+
step 140000 | 1450s | train_loss=0.2683
|
| 17 |
+
step 150000 | 1544s | train_loss=0.2257
|
| 18 |
+
step 160000 | 1639s | train_loss=0.1978
|
| 19 |
+
step 170000 | 1733s | train_loss=0.1823
|
| 20 |
+
step 180000 | 1827s | train_loss=0.1672
|
| 21 |
+
step 190000 | 1921s | train_loss=0.1549
|
| 22 |
+
step 200000 | 2081s | train_loss=0.1504 | fid=163.7400
|
| 23 |
+
step 210000 | 2174s | train_loss=0.1404
|
| 24 |
+
step 220000 | 2268s | train_loss=0.1329
|
| 25 |
+
step 230000 | 2362s | train_loss=0.1253
|
| 26 |
+
step 240000 | 2456s | train_loss=0.1234
|
| 27 |
+
step 250000 | 2550s | train_loss=0.1171
|
| 28 |
+
step 260000 | 2643s | train_loss=0.1115
|
| 29 |
+
step 270000 | 2737s | train_loss=0.1063
|
| 30 |
+
step 280000 | 2831s | train_loss=0.1041
|
| 31 |
+
step 290000 | 2919s | train_loss=0.1002
|
| 32 |
+
step 300000 | 3085s | train_loss=0.0965 | fid=163.5126
|
| 33 |
+
step 310000 | 3179s | train_loss=0.0927
|
| 34 |
+
step 320000 | 3272s | train_loss=0.0910
|
| 35 |
+
step 330000 | 3366s | train_loss=0.0885
|
| 36 |
+
step 340000 | 3460s | train_loss=0.0854
|
| 37 |
+
step 350000 | 3554s | train_loss=0.0851
|
| 38 |
+
step 360000 | 3648s | train_loss=0.0822
|
| 39 |
+
step 370000 | 3742s | train_loss=0.0797
|
| 40 |
+
step 380000 | 3833s | train_loss=0.0772
|
| 41 |
+
step 390000 | 3924s | train_loss=0.0756
|
| 42 |
+
step 400000 | 4089s | train_loss=0.0737 | fid=160.9712
|
| 43 |
+
step 410000 | 4183s | train_loss=0.0726
|
| 44 |
+
step 420000 | 4276s | train_loss=0.0705
|
| 45 |
+
step 430000 | 4370s | train_loss=0.0686
|
| 46 |
+
step 440000 | 4464s | train_loss=0.0689
|
| 47 |
+
step 450000 | 4558s | train_loss=0.0673
|
| 48 |
+
step 460000 | 4651s | train_loss=0.0654
|
| 49 |
+
step 470000 | 4742s | train_loss=0.0647
|
| 50 |
+
step 480000 | 4833s | train_loss=0.0630
|
| 51 |
+
step 490000 | 4927s | train_loss=0.0622
|
| 52 |
+
step 500000 | 5094s | train_loss=0.0609 | fid=159.9465
|
| 53 |
+
step 510000 | 5187s | train_loss=0.0597
|
| 54 |
+
step 520000 | 5279s | train_loss=0.0586
|
| 55 |
+
step 530000 | 5374s | train_loss=0.0579
|
| 56 |
+
step 540000 | 5468s | train_loss=0.0570
|
| 57 |
+
step 550000 | 5563s | train_loss=0.0551
|
| 58 |
+
step 560000 | 5657s | train_loss=0.0552
|
| 59 |
+
step 570000 | 5745s | train_loss=0.0549
|
| 60 |
+
step 580000 | 5839s | train_loss=0.0537
|
| 61 |
+
step 590000 | 5933s | train_loss=0.0535
|
| 62 |
+
step 600000 | 6099s | train_loss=0.0524 | fid=159.3085
|
| 63 |
+
step 610000 | 6193s | train_loss=0.0509
|
| 64 |
+
step 620000 | 6287s | train_loss=0.0505
|
| 65 |
+
step 630000 | 6380s | train_loss=0.0503
|
| 66 |
+
step 640000 | 6474s | train_loss=0.0493
|
| 67 |
+
step 650000 | 6568s | train_loss=0.0494
|
| 68 |
+
step 660000 | 6659s | train_loss=0.0483
|
| 69 |
+
step 670000 | 6750s | train_loss=0.0477
|
| 70 |
+
step 680000 | 6844s | train_loss=0.0460
|
| 71 |
+
step 690000 | 6938s | train_loss=0.0468
|
| 72 |
+
step 700000 | 7104s | train_loss=0.0463 | fid=158.3443
|
| 73 |
+
step 710000 | 7198s | train_loss=0.0454
|
| 74 |
+
step 720000 | 7292s | train_loss=0.0442
|
| 75 |
+
step 730000 | 7385s | train_loss=0.0444
|
| 76 |
+
step 740000 | 7479s | train_loss=0.0439
|
| 77 |
+
step 750000 | 7570s | train_loss=0.0438
|
| 78 |
+
step 760000 | 7660s | train_loss=0.0431
|
| 79 |
+
step 770000 | 7755s | train_loss=0.0438
|
| 80 |
+
step 780000 | 7849s | train_loss=0.0423
|
| 81 |
+
step 790000 | 7943s | train_loss=0.0420
|
| 82 |
+
step 800000 | 8108s | train_loss=0.0410 | fid=157.5954
|
| 83 |
+
step 810000 | 8202s | train_loss=0.0408
|
| 84 |
+
step 820000 | 8296s | train_loss=0.0402
|
| 85 |
+
step 830000 | 8390s | train_loss=0.0416
|
| 86 |
+
step 840000 | 8480s | train_loss=0.0396
|
| 87 |
+
step 850000 | 8571s | train_loss=0.0396
|
| 88 |
+
step 860000 | 8666s | train_loss=0.0393
|
| 89 |
+
step 870000 | 8760s | train_loss=0.0382
|
| 90 |
+
step 880000 | 8854s | train_loss=0.0391
|
| 91 |
+
step 890000 | 8948s | train_loss=0.0378
|
| 92 |
+
step 900000 | 9114s | train_loss=0.0380 | fid=156.0037
|
| 93 |
+
step 910000 | 9207s | train_loss=0.0381
|
| 94 |
+
step 920000 | 9302s | train_loss=0.0374
|
| 95 |
+
step 930000 | 9396s | train_loss=0.0370
|
| 96 |
+
step 940000 | 9485s | train_loss=0.0368
|
| 97 |
+
step 950000 | 9576s | train_loss=0.0365
|
| 98 |
+
step 960000 | 9670s | train_loss=0.0362
|
| 99 |
+
step 970000 | 9764s | train_loss=0.0367
|
| 100 |
+
step 980000 | 9859s | train_loss=0.0351
|
| 101 |
+
step 990000 | 9953s | train_loss=0.0361
|
| 102 |
+
step 1000000 | 10116s | train_loss=0.0356 | fid=156.7306
|
| 103 |
+
step 1010000 | 10210s | train_loss=0.0347
|
| 104 |
+
step 1020000 | 10304s | train_loss=0.0348
|
| 105 |
+
step 1030000 | 10394s | train_loss=0.0344
|
| 106 |
+
step 1040000 | 10483s | train_loss=0.0341
|
| 107 |
+
step 1050000 | 10577s | train_loss=0.0345
|
| 108 |
+
step 1060000 | 10672s | train_loss=0.0343
|
| 109 |
+
step 1070000 | 10766s | train_loss=0.0334
|
| 110 |
+
step 1080000 | 10860s | train_loss=0.0329
|
| 111 |
+
step 1090000 | 10955s | train_loss=0.0334
|
| 112 |
+
step 1100000 | 11119s | train_loss=0.0322 | fid=156.6190
|
| 113 |
+
step 1110000 | 11213s | train_loss=0.0326
|
| 114 |
+
step 1120000 | 11304s | train_loss=0.0322
|
| 115 |
+
step 1130000 | 11392s | train_loss=0.0324
|
| 116 |
+
step 1140000 | 11479s | train_loss=0.0317
|
| 117 |
+
step 1150000 | 11571s | train_loss=0.0321
|
| 118 |
+
step 1160000 | 11665s | train_loss=0.0321
|
| 119 |
+
step 1170000 | 11760s | train_loss=0.0311
|
| 120 |
+
step 1180000 | 11854s | train_loss=0.0307
|
| 121 |
+
step 1190000 | 11948s | train_loss=0.0305
|
| 122 |
+
step 1200000 | 12112s | train_loss=0.0309 | fid=156.0789
|
| 123 |
+
step 1210000 | 12204s | train_loss=0.0313
|
| 124 |
+
step 1220000 | 12298s | train_loss=0.0307
|
| 125 |
+
step 1230000 | 12390s | train_loss=0.0308
|
| 126 |
+
step 1240000 | 12484s | train_loss=0.0305
|
| 127 |
+
step 1250000 | 12576s | train_loss=0.0301
|
| 128 |
+
step 1260000 | 12670s | train_loss=0.0303
|
| 129 |
+
step 1270000 | 12764s | train_loss=0.0298
|
| 130 |
+
step 1280000 | 12858s | train_loss=0.0298
|
| 131 |
+
step 1290000 | 12952s | train_loss=0.0291
|
| 132 |
+
step 1300000 | 13116s | train_loss=0.0296 | fid=155.4446
|
| 133 |
+
step 1310000 | 13194s | train_loss=0.0294
|
| 134 |
+
step 1320000 | 13286s | train_loss=0.0293
|
| 135 |
+
step 1330000 | 13368s | train_loss=0.0287
|
| 136 |
+
step 1340000 | 13463s | train_loss=0.0288
|
| 137 |
+
step 1350000 | 13555s | train_loss=0.0289
|
| 138 |
+
step 1360000 | 13649s | train_loss=0.0287
|
| 139 |
+
step 1370000 | 13743s | train_loss=0.0287
|
| 140 |
+
step 1380000 | 13837s | train_loss=0.0278
|
| 141 |
+
step 1390000 | 13931s | train_loss=0.0281
|
| 142 |
+
step 1400000 | 14088s | train_loss=0.0277 | fid=155.1377
|
| 143 |
+
step 1410000 | 14173s | train_loss=0.0284
|
| 144 |
+
step 1420000 | 14263s | train_loss=0.0287
|
| 145 |
+
step 1430000 | 14358s | train_loss=0.0276
|
| 146 |
+
step 1440000 | 14452s | train_loss=0.0276
|
| 147 |
+
step 1450000 | 14546s | train_loss=0.0270
|
| 148 |
+
step 1460000 | 14639s | train_loss=0.0273
|
| 149 |
+
step 1470000 | 14733s | train_loss=0.0271
|
| 150 |
+
step 1480000 | 14827s | train_loss=0.0275
|
| 151 |
+
step 1490000 | 14921s | train_loss=0.0270
|
| 152 |
+
step 1500000 | 15079s | train_loss=0.0264 | fid=155.7748
|
| 153 |
+
step 1510000 | 15173s | train_loss=0.0263
|
| 154 |
+
step 1520000 | 15265s | train_loss=0.0266
|
| 155 |
+
step 1530000 | 15359s | train_loss=0.0261
|
| 156 |
+
step 1540000 | 15453s | train_loss=0.0263
|
| 157 |
+
step 1550000 | 15547s | train_loss=0.0262
|
| 158 |
+
step 1560000 | 15640s | train_loss=0.0265
|
| 159 |
+
step 1570000 | 15734s | train_loss=0.0258
|
| 160 |
+
step 1580000 | 15828s | train_loss=0.0257
|
| 161 |
+
step 1590000 | 15922s | train_loss=0.0252
|
| 162 |
+
step 1600000 | 16084s | train_loss=0.0262 | fid=155.6058
|
| 163 |
+
step 1610000 | 16169s | train_loss=0.0253
|
| 164 |
+
step 1620000 | 16264s | train_loss=0.0256
|
| 165 |
+
step 1630000 | 16358s | train_loss=0.0250
|
| 166 |
+
step 1640000 | 16452s | train_loss=0.0251
|
| 167 |
+
step 1650000 | 16547s | train_loss=0.0249
|
| 168 |
+
step 1660000 | 16641s | train_loss=0.0246
|
| 169 |
+
step 1670000 | 16733s | train_loss=0.0244
|
| 170 |
+
step 1680000 | 16827s | train_loss=0.0247
|
| 171 |
+
step 1690000 | 16920s | train_loss=0.0246
|
| 172 |
+
step 1700000 | 17067s | train_loss=0.0243 | fid=155.3769
|
| 173 |
+
step 1710000 | 17158s | train_loss=0.0244
|
| 174 |
+
step 1720000 | 17252s | train_loss=0.0246
|
| 175 |
+
step 1730000 | 17346s | train_loss=0.0241
|
| 176 |
+
step 1740000 | 17441s | train_loss=0.0237
|
| 177 |
+
step 1750000 | 17535s | train_loss=0.0244
|
| 178 |
+
step 1760000 | 17629s | train_loss=0.0239
|
| 179 |
+
step 1770000 | 17722s | train_loss=0.0239
|
| 180 |
+
step 1780000 | 17816s | train_loss=0.0242
|
| 181 |
+
step 1790000 | 17907s | train_loss=0.0239
|
| 182 |
+
step 1800000 | 18067s | train_loss=0.0235 | fid=153.8409
|
| 183 |
+
step 1810000 | 18161s | train_loss=0.0232
|
| 184 |
+
step 1820000 | 18255s | train_loss=0.0235
|
| 185 |
+
step 1830000 | 18349s | train_loss=0.0236
|
| 186 |
+
step 1840000 | 18444s | train_loss=0.0230
|
| 187 |
+
step 1850000 | 18538s | train_loss=0.0229
|
| 188 |
+
step 1860000 | 18632s | train_loss=0.0230
|
| 189 |
+
step 1870000 | 18725s | train_loss=0.0232
|
| 190 |
+
step 1880000 | 18819s | train_loss=0.0226
|
| 191 |
+
step 1890000 | 18911s | train_loss=0.0231
|
| 192 |
+
step 1900000 | 19072s | train_loss=0.0225 | fid=154.7188
|
| 193 |
+
step 1910000 | 19159s | train_loss=0.0229
|
| 194 |
+
step 1920000 | 19253s | train_loss=0.0227
|
| 195 |
+
step 1930000 | 19347s | train_loss=0.0228
|
| 196 |
+
step 1940000 | 19441s | train_loss=0.0221
|
| 197 |
+
step 1950000 | 19535s | train_loss=0.0222
|
| 198 |
+
step 1960000 | 19629s | train_loss=0.0222
|
| 199 |
+
step 1970000 | 19723s | train_loss=0.0217
|
| 200 |
+
step 1980000 | 19801s | train_loss=0.0223
|
| 201 |
+
step 1990000 | 19893s | train_loss=0.0221
|
| 202 |
+
step 2000000 | 20047s | train_loss=0.0215 | fid=155.6034
|
| 203 |
+
step 2010000 | 20132s | train_loss=0.0219
|
| 204 |
+
step 2020000 | 20226s | train_loss=0.0221
|
| 205 |
+
step 2030000 | 20319s | train_loss=0.0217
|
| 206 |
+
step 2040000 | 20413s | train_loss=0.0214
|
| 207 |
+
step 2050000 | 20508s | train_loss=0.0215
|
| 208 |
+
step 2060000 | 20602s | train_loss=0.0215
|
| 209 |
+
step 2070000 | 20696s | train_loss=0.0214
|
| 210 |
+
step 2080000 | 20771s | train_loss=0.0217
|
| 211 |
+
step 2090000 | 20863s | train_loss=0.0210
|
| 212 |
+
step 2100000 | 21015s | train_loss=0.0208 | fid=154.5192
|
| 213 |
+
step 2110000 | 21101s | train_loss=0.0214
|
| 214 |
+
step 2120000 | 21189s | train_loss=0.0213
|
| 215 |
+
step 2130000 | 21284s | train_loss=0.0205
|
| 216 |
+
step 2140000 | 21377s | train_loss=0.0206
|
| 217 |
+
step 2150000 | 21471s | train_loss=0.0209
|
| 218 |
+
step 2160000 | 21566s | train_loss=0.0213
|
| 219 |
+
step 2170000 | 21660s | train_loss=0.0213
|
| 220 |
+
step 2180000 | 21739s | train_loss=0.0207
|
| 221 |
+
step 2190000 | 21831s | train_loss=0.0206
|
| 222 |
+
step 2200000 | 21988s | train_loss=0.0202 | fid=155.7955
|
| 223 |
+
step 2210000 | 22082s | train_loss=0.0208
|
| 224 |
+
step 2220000 | 22162s | train_loss=0.0204
|
| 225 |
+
step 2230000 | 22256s | train_loss=0.0208
|
| 226 |
+
step 2240000 | 22350s | train_loss=0.0203
|
| 227 |
+
step 2250000 | 22443s | train_loss=0.0204
|
| 228 |
+
step 2260000 | 22538s | train_loss=0.0202
|
| 229 |
+
step 2270000 | 22632s | train_loss=0.0200
|
| 230 |
+
step 2280000 | 22713s | train_loss=0.0202
|
| 231 |
+
step 2290000 | 22801s | train_loss=0.0204
|
| 232 |
+
step 2300000 | 22956s | train_loss=0.0197 | fid=155.2028
|
| 233 |
+
step 2310000 | 23051s | train_loss=0.0196
|
| 234 |
+
step 2320000 | 23138s | train_loss=0.0201
|
| 235 |
+
step 2330000 | 23227s | train_loss=0.0197
|
| 236 |
+
step 2340000 | 23321s | train_loss=0.0202
|
| 237 |
+
step 2350000 | 23415s | train_loss=0.0198
|
| 238 |
+
step 2360000 | 23510s | train_loss=0.0196
|
| 239 |
+
step 2370000 | 23600s | train_loss=0.0195
|
| 240 |
+
step 2380000 | 23682s | train_loss=0.0198
|
| 241 |
+
step 2390000 | 23770s | train_loss=0.0190
|
| 242 |
+
step 2400000 | 23930s | train_loss=0.0194 | fid=155.1195
|
| 243 |
+
step 2410000 | 24024s | train_loss=0.0193
|
| 244 |
+
step 2420000 | 24118s | train_loss=0.0191
|
| 245 |
+
step 2430000 | 24195s | train_loss=0.0192
|
| 246 |
+
step 2440000 | 24288s | train_loss=0.0189
|
| 247 |
+
step 2450000 | 24382s | train_loss=0.0192
|
| 248 |
+
step 2460000 | 24476s | train_loss=0.0189
|
| 249 |
+
step 2470000 | 24566s | train_loss=0.0187
|
| 250 |
+
step 2480000 | 24652s | train_loss=0.0188
|
| 251 |
+
step 2490000 | 24739s | train_loss=0.0184
|
| 252 |
+
step 2500000 | 24900s | train_loss=0.0185 | fid=154.5770
|
| 253 |
+
step 2510000 | 24994s | train_loss=0.0188
|
| 254 |
+
step 2520000 | 25088s | train_loss=0.0195
|
| 255 |
+
step 2530000 | 25180s | train_loss=0.0184
|
| 256 |
+
step 2540000 | 25261s | train_loss=0.0183
|
| 257 |
+
step 2550000 | 25356s | train_loss=0.0185
|
| 258 |
+
step 2560000 | 25450s | train_loss=0.0185
|
| 259 |
+
step 2570000 | 25536s | train_loss=0.0185
|
| 260 |
+
step 2580000 | 25624s | train_loss=0.0187
|
| 261 |
+
step 2590000 | 25707s | train_loss=0.0186
|
| 262 |
+
step 2600000 | 25872s | train_loss=0.0180 | fid=154.8491
|
| 263 |
+
step 2610000 | 25966s | train_loss=0.0179
|
| 264 |
+
step 2620000 | 26060s | train_loss=0.0179
|
| 265 |
+
step 2630000 | 26154s | train_loss=0.0182
|
| 266 |
+
step 2640000 | 26237s | train_loss=0.0184
|
| 267 |
+
step 2650000 | 26326s | train_loss=0.0181
|
| 268 |
+
step 2660000 | 26420s | train_loss=0.0175
|
| 269 |
+
step 2670000 | 26504s | train_loss=0.0181
|
| 270 |
+
step 2680000 | 26598s | train_loss=0.0178
|
| 271 |
+
step 2690000 | 26684s | train_loss=0.0175
|
| 272 |
+
step 2700000 | 26849s | train_loss=0.0176 | fid=155.8501
|
| 273 |
+
step 2710000 | 26944s | train_loss=0.0179
|
| 274 |
+
step 2720000 | 27038s | train_loss=0.0178
|
| 275 |
+
step 2730000 | 27132s | train_loss=0.0178
|
| 276 |
+
step 2740000 | 27226s | train_loss=0.0182
|
| 277 |
+
step 2750000 | 27304s | train_loss=0.0176
|
| 278 |
+
step 2760000 | 27398s | train_loss=0.0172
|
| 279 |
+
step 2770000 | 27478s | train_loss=0.0177
|
| 280 |
+
step 2780000 | 27570s | train_loss=0.0172
|
| 281 |
+
step 2790000 | 27648s | train_loss=0.0172
|
| 282 |
+
step 2800000 | 27814s | train_loss=0.0172 | fid=154.5480
|
| 283 |
+
step 2810000 | 27908s | train_loss=0.0173
|
| 284 |
+
step 2820000 | 28003s | train_loss=0.0171
|
| 285 |
+
step 2830000 | 28096s | train_loss=0.0174
|
| 286 |
+
step 2840000 | 28190s | train_loss=0.0175
|
| 287 |
+
step 2850000 | 28281s | train_loss=0.0171
|
| 288 |
+
step 2860000 | 28365s | train_loss=0.0170
|
| 289 |
+
step 2870000 | 28451s | train_loss=0.0169
|
| 290 |
+
step 2880000 | 28545s | train_loss=0.0169
|
| 291 |
+
step 2890000 | 28624s | train_loss=0.0170
|
| 292 |
+
step 2900000 | 28791s | train_loss=0.0166 | fid=155.6896
|
| 293 |
+
step 2910000 | 28883s | train_loss=0.0164
|
| 294 |
+
step 2920000 | 28978s | train_loss=0.0168
|
| 295 |
+
step 2930000 | 29072s | train_loss=0.0165
|
| 296 |
+
step 2940000 | 29166s | train_loss=0.0166
|
| 297 |
+
step 2950000 | 29261s | train_loss=0.0164
|
| 298 |
+
step 2960000 | 29333s | train_loss=0.0162
|
| 299 |
+
step 2970000 | 29423s | train_loss=0.0169
|
| 300 |
+
step 2980000 | 29518s | train_loss=0.0165
|
| 301 |
+
step 2990000 | 29607s | train_loss=0.0161
|
| 302 |
+
step 3000000 | 29775s | train_loss=0.0165 | fid=154.7794
|
| 303 |
+
step 3010000 | 29867s | train_loss=0.0168
|
| 304 |
+
step 3020000 | 29961s | train_loss=0.0166
|
| 305 |
+
step 3030000 | 30055s | train_loss=0.0160
|
| 306 |
+
step 3040000 | 30148s | train_loss=0.0167
|
| 307 |
+
step 3050000 | 30242s | train_loss=0.0165
|
| 308 |
+
step 3060000 | 30319s | train_loss=0.0159
|
| 309 |
+
step 3070000 | 30403s | train_loss=0.0164
|
| 310 |
+
step 3080000 | 30488s | train_loss=0.0165
|
| 311 |
+
step 3090000 | 30577s | train_loss=0.0162
|
| 312 |
+
step 3100000 | 30744s | train_loss=0.0163 | fid=155.0566
|
| 313 |
+
step 3110000 | 30837s | train_loss=0.0159
|
| 314 |
+
step 3120000 | 30931s | train_loss=0.0163
|
| 315 |
+
step 3130000 | 31025s | train_loss=0.0157
|
| 316 |
+
step 3140000 | 31118s | train_loss=0.0158
|
| 317 |
+
step 3150000 | 31212s | train_loss=0.0157
|
| 318 |
+
step 3160000 | 31296s | train_loss=0.0154
|
| 319 |
+
step 3170000 | 31379s | train_loss=0.0160
|
| 320 |
+
step 3180000 | 31464s | train_loss=0.0158
|
| 321 |
+
step 3190000 | 31554s | train_loss=0.0154
|
| 322 |
+
step 3200000 | 31722s | train_loss=0.0159 | fid=155.5518
|
| 323 |
+
step 3210000 | 31816s | train_loss=0.0159
|
| 324 |
+
step 3220000 | 31909s | train_loss=0.0159
|
| 325 |
+
step 3230000 | 32003s | train_loss=0.0154
|
| 326 |
+
step 3240000 | 32097s | train_loss=0.0155
|
| 327 |
+
step 3250000 | 32185s | train_loss=0.0156
|
| 328 |
+
step 3260000 | 32272s | train_loss=0.0154
|
| 329 |
+
step 3270000 | 32363s | train_loss=0.0157
|
| 330 |
+
step 3280000 | 32438s | train_loss=0.0155
|
| 331 |
+
step 3290000 | 32530s | train_loss=0.0153
|
| 332 |
+
step 3300000 | 32698s | train_loss=0.0157 | fid=154.7110
|
| 333 |
+
step 3310000 | 32792s | train_loss=0.0151
|
| 334 |
+
step 3320000 | 32884s | train_loss=0.0151
|
| 335 |
+
step 3330000 | 32978s | train_loss=0.0150
|
| 336 |
+
step 3340000 | 33072s | train_loss=0.0152
|
| 337 |
+
step 3350000 | 33159s | train_loss=0.0148
|
| 338 |
+
step 3360000 | 33253s | train_loss=0.0152
|
| 339 |
+
step 3370000 | 33348s | train_loss=0.0153
|
| 340 |
+
step 3380000 | 33419s | train_loss=0.0149
|
| 341 |
+
step 3390000 | 33513s | train_loss=0.0147
|
| 342 |
+
step 3400000 | 33680s | train_loss=0.0149 | fid=155.2607
|
| 343 |
+
step 3410000 | 33773s | train_loss=0.0152
|
| 344 |
+
step 3420000 | 33867s | train_loss=0.0146
|
| 345 |
+
step 3430000 | 33960s | train_loss=0.0157
|
| 346 |
+
step 3440000 | 34052s | train_loss=0.0152
|
| 347 |
+
step 3450000 | 34141s | train_loss=0.0151
|
| 348 |
+
step 3460000 | 34235s | train_loss=0.0149
|
| 349 |
+
step 3470000 | 34330s | train_loss=0.0147
|
| 350 |
+
step 3480000 | 34408s | train_loss=0.0148
|
| 351 |
+
step 3490000 | 34496s | train_loss=0.0145
|
| 352 |
+
step 3500000 | 34663s | train_loss=0.0147 | fid=155.8554
|
| 353 |
+
step 3510000 | 34758s | train_loss=0.0145
|
| 354 |
+
step 3520000 | 34852s | train_loss=0.0146
|
| 355 |
+
step 3530000 | 34944s | train_loss=0.0149
|
| 356 |
+
step 3540000 | 35026s | train_loss=0.0147
|
| 357 |
+
step 3550000 | 35119s | train_loss=0.0145
|
| 358 |
+
step 3560000 | 35213s | train_loss=0.0145
|
| 359 |
+
step 3570000 | 35303s | train_loss=0.0141
|
| 360 |
+
step 3580000 | 35384s | train_loss=0.0143
|
| 361 |
+
step 3590000 | 35462s | train_loss=0.0144
|
| 362 |
+
step 3600000 | 35630s | train_loss=0.0148 | fid=155.4899
|
| 363 |
+
step 3610000 | 35724s | train_loss=0.0141
|
| 364 |
+
step 3620000 | 35818s | train_loss=0.0146
|
| 365 |
+
step 3630000 | 35910s | train_loss=0.0142
|
| 366 |
+
step 3640000 | 35989s | train_loss=0.0144
|
| 367 |
+
step 3650000 | 36084s | train_loss=0.0145
|
| 368 |
+
step 3660000 | 36178s | train_loss=0.0142
|
| 369 |
+
step 3670000 | 36271s | train_loss=0.0142
|
| 370 |
+
step 3680000 | 36356s | train_loss=0.0143
|
| 371 |
+
step 3690000 | 36446s | train_loss=0.0143
|
| 372 |
+
step 3700000 | 36608s | train_loss=0.0144 | fid=155.5126
|
| 373 |
+
step 3710000 | 36702s | train_loss=0.0143
|
| 374 |
+
step 3720000 | 36796s | train_loss=0.0141
|
| 375 |
+
step 3730000 | 36890s | train_loss=0.0144
|
| 376 |
+
step 3740000 | 36971s | train_loss=0.0140
|
| 377 |
+
step 3750000 | 37065s | train_loss=0.0142
|
| 378 |
+
step 3760000 | 37159s | train_loss=0.0138
|
| 379 |
+
step 3770000 | 37247s | train_loss=0.0135
|
| 380 |
+
step 3780000 | 37336s | train_loss=0.0137
|
| 381 |
+
step 3790000 | 37430s | train_loss=0.0138
|
| 382 |
+
step 3800000 | 37583s | train_loss=0.0138 | fid=154.7989
|
| 383 |
+
step 3810000 | 37678s | train_loss=0.0141
|
| 384 |
+
step 3820000 | 37772s | train_loss=0.0136
|
| 385 |
+
step 3830000 | 37860s | train_loss=0.0135
|
| 386 |
+
step 3840000 | 37944s | train_loss=0.0141
|
| 387 |
+
step 3850000 | 38038s | train_loss=0.0136
|
| 388 |
+
step 3860000 | 38132s | train_loss=0.0137
|
| 389 |
+
step 3870000 | 38221s | train_loss=0.0135
|
| 390 |
+
step 3880000 | 38315s | train_loss=0.0133
|
| 391 |
+
step 3890000 | 38409s | train_loss=0.0136
|
| 392 |
+
step 3900000 | 38574s | train_loss=0.0137 | fid=154.6547
|
| 393 |
+
step 3910000 | 38668s | train_loss=0.0136
|
| 394 |
+
step 3920000 | 38763s | train_loss=0.0137
|
| 395 |
+
step 3930000 | 38854s | train_loss=0.0137
|
| 396 |
+
step 3940000 | 38949s | train_loss=0.0136
|
| 397 |
+
step 3950000 | 39041s | train_loss=0.0135
|
| 398 |
+
step 3960000 | 39134s | train_loss=0.0135
|
| 399 |
+
step 3970000 | 39226s | train_loss=0.0136
|
| 400 |
+
step 3980000 | 39321s | train_loss=0.0136
|
| 401 |
+
step 3990000 | 39415s | train_loss=0.0134
|
| 402 |
+
step 4000000 | 39579s | train_loss=0.0137 | fid=155.0101
|
| 403 |
+
step 4010000 | 39673s | train_loss=0.0135
|
| 404 |
+
step 4020000 | 39765s | train_loss=0.0133
|
| 405 |
+
step 4030000 | 39860s | train_loss=0.0134
|
| 406 |
+
step 4040000 | 39954s | train_loss=0.0134
|
| 407 |
+
step 4050000 | 40047s | train_loss=0.0134
|
| 408 |
+
step 4060000 | 40138s | train_loss=0.0131
|
| 409 |
+
step 4070000 | 40232s | train_loss=0.0133
|
| 410 |
+
step 4080000 | 40326s | train_loss=0.0129
|
| 411 |
+
step 4090000 | 40421s | train_loss=0.0133
|
| 412 |
+
step 4100000 | 40579s | train_loss=0.0132 | fid=155.4590
|
| 413 |
+
step 4110000 | 40673s | train_loss=0.0136
|
| 414 |
+
step 4120000 | 40750s | train_loss=0.0132
|
| 415 |
+
step 4130000 | 40844s | train_loss=0.0133
|
| 416 |
+
step 4140000 | 40938s | train_loss=0.0132
|
| 417 |
+
step 4150000 | 41022s | train_loss=0.0133
|
| 418 |
+
step 4160000 | 41108s | train_loss=0.0130
|
| 419 |
+
step 4170000 | 41202s | train_loss=0.0130
|
| 420 |
+
step 4180000 | 41296s | train_loss=0.0130
|
| 421 |
+
step 4190000 | 41390s | train_loss=0.0133
|
| 422 |
+
step 4200000 | 41547s | train_loss=0.0129 | fid=155.7969
|
| 423 |
+
step 4210000 | 41641s | train_loss=0.0129
|
| 424 |
+
step 4220000 | 41725s | train_loss=0.0128
|
| 425 |
+
step 4230000 | 41819s | train_loss=0.0129
|
| 426 |
+
step 4240000 | 41913s | train_loss=0.0129
|
| 427 |
+
step 4250000 | 42001s | train_loss=0.0128
|
| 428 |
+
step 4260000 | 42086s | train_loss=0.0127
|
| 429 |
+
step 4270000 | 42180s | train_loss=0.0131
|
| 430 |
+
step 4280000 | 42274s | train_loss=0.0128
|
| 431 |
+
step 4290000 | 42368s | train_loss=0.0129
|
| 432 |
+
step 4300000 | 42528s | train_loss=0.0126 | fid=154.5375
|
| 433 |
+
step 4310000 | 42612s | train_loss=0.0128
|
| 434 |
+
step 4320000 | 42697s | train_loss=0.0126
|
| 435 |
+
step 4330000 | 42790s | train_loss=0.0125
|
| 436 |
+
step 4340000 | 42885s | train_loss=0.0126
|
| 437 |
+
step 4350000 | 42970s | train_loss=0.0127
|
| 438 |
+
step 4360000 | 43055s | train_loss=0.0128
|
| 439 |
+
step 4370000 | 43149s | train_loss=0.0126
|
| 440 |
+
step 4380000 | 43244s | train_loss=0.0128
|
| 441 |
+
step 4390000 | 43338s | train_loss=0.0130
|
| 442 |
+
step 4400000 | 43505s | train_loss=0.0129 | fid=155.4211
|
| 443 |
+
step 4410000 | 43578s | train_loss=0.0129
|
| 444 |
+
step 4420000 | 43666s | train_loss=0.0126
|
| 445 |
+
step 4430000 | 43761s | train_loss=0.0123
|
| 446 |
+
step 4440000 | 43855s | train_loss=0.0125
|
| 447 |
+
step 4450000 | 43941s | train_loss=0.0124
|
| 448 |
+
step 4460000 | 44034s | train_loss=0.0123
|
| 449 |
+
step 4470000 | 44126s | train_loss=0.0128
|
| 450 |
+
step 4480000 | 44219s | train_loss=0.0127
|
| 451 |
+
step 4490000 | 44314s | train_loss=0.0125
|
| 452 |
+
step 4500000 | 44481s | train_loss=0.0123 | fid=154.4054
|
| 453 |
+
step 4510000 | 44556s | train_loss=0.0126
|
| 454 |
+
step 4520000 | 44646s | train_loss=0.0124
|
| 455 |
+
step 4530000 | 44740s | train_loss=0.0126
|
| 456 |
+
step 4540000 | 44834s | train_loss=0.0127
|
| 457 |
+
step 4550000 | 44914s | train_loss=0.0120
|
| 458 |
+
step 4560000 | 45008s | train_loss=0.0122
|
| 459 |
+
step 4570000 | 45100s | train_loss=0.0122
|
| 460 |
+
step 4580000 | 45194s | train_loss=0.0123
|
| 461 |
+
step 4590000 | 45289s | train_loss=0.0124
|
| 462 |
+
step 4600000 | 45456s | train_loss=0.0119 | fid=155.3044
|
| 463 |
+
step 4610000 | 45537s | train_loss=0.0120
|
| 464 |
+
step 4620000 | 45612s | train_loss=0.0122
|
| 465 |
+
step 4630000 | 45706s | train_loss=0.0122
|
| 466 |
+
step 4640000 | 45801s | train_loss=0.0125
|
| 467 |
+
step 4650000 | 45882s | train_loss=0.0120
|
| 468 |
+
step 4660000 | 45976s | train_loss=0.0124
|
| 469 |
+
step 4670000 | 46068s | train_loss=0.0123
|
| 470 |
+
step 4680000 | 46162s | train_loss=0.0121
|
| 471 |
+
step 4690000 | 46256s | train_loss=0.0122
|
| 472 |
+
step 4700000 | 46424s | train_loss=0.0119 | fid=154.7513
|
| 473 |
+
step 4710000 | 46507s | train_loss=0.0120
|
| 474 |
+
step 4720000 | 46597s | train_loss=0.0118
|
| 475 |
+
step 4730000 | 46681s | train_loss=0.0119
|
| 476 |
+
step 4740000 | 46776s | train_loss=0.0119
|
| 477 |
+
step 4750000 | 46861s | train_loss=0.0118
|
| 478 |
+
step 4760000 | 46955s | train_loss=0.0120
|
| 479 |
+
step 4770000 | 47049s | train_loss=0.0119
|
| 480 |
+
step 4780000 | 47141s | train_loss=0.0116
|
| 481 |
+
step 4790000 | 47235s | train_loss=0.0118
|
| 482 |
+
step 4800000 | 47399s | train_loss=0.0117 | fid=155.3522
|
| 483 |
+
step 4810000 | 47475s | train_loss=0.0117
|
| 484 |
+
step 4820000 | 47551s | train_loss=0.0118
|
| 485 |
+
step 4830000 | 47621s | train_loss=0.0117
|
| 486 |
+
step 4840000 | 47692s | train_loss=0.0121
|
| 487 |
+
step 4850000 | 47751s | train_loss=0.0118
|
| 488 |
+
step 4860000 | 47809s | train_loss=0.0116
|
| 489 |
+
step 4870000 | 47867s | train_loss=0.0118
|
| 490 |
+
step 4880000 | 47922s | train_loss=0.0117
|
| 491 |
+
step 4890000 | 47980s | train_loss=0.0120
|
| 492 |
+
step 4900000 | 48086s | train_loss=0.0116 | fid=154.9541
|
| 493 |
+
step 4910000 | 48144s | train_loss=0.0120
|
| 494 |
+
step 4920000 | 48203s | train_loss=0.0116
|
| 495 |
+
step 4930000 | 48255s | train_loss=0.0120
|
| 496 |
+
step 4940000 | 48313s | train_loss=0.0117
|
| 497 |
+
step 4950000 | 48372s | train_loss=0.0117
|
| 498 |
+
step 4960000 | 48430s | train_loss=0.0119
|
| 499 |
+
step 4970000 | 48488s | train_loss=0.0116
|
| 500 |
+
step 4980000 | 48546s | train_loss=0.0117
|
| 501 |
+
step 4990000 | 48601s | train_loss=0.0115
|
| 502 |
+
step 5000000 | 48708s | train_loss=0.0115 | fid=155.0162
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d200/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d200/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d200",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 44,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d200/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d200/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 86s | train_loss=1.0040 | fid=298.6502
|
| 3 |
+
step 10000 | 195s | train_loss=0.7375
|
| 4 |
+
step 20000 | 305s | train_loss=0.6225
|
| 5 |
+
step 30000 | 415s | train_loss=0.5441
|
| 6 |
+
step 40000 | 526s | train_loss=0.4773
|
| 7 |
+
step 50000 | 636s | train_loss=0.4058
|
| 8 |
+
step 60000 | 746s | train_loss=0.3371
|
| 9 |
+
step 70000 | 856s | train_loss=0.2834
|
| 10 |
+
step 80000 | 967s | train_loss=0.2373
|
| 11 |
+
step 90000 | 1077s | train_loss=0.2110
|
| 12 |
+
step 100000 | 1269s | train_loss=0.1798 | fid=182.9636
|
| 13 |
+
step 110000 | 1375s | train_loss=0.1586
|
| 14 |
+
step 120000 | 1484s | train_loss=0.1509
|
| 15 |
+
step 130000 | 1594s | train_loss=0.1450
|
| 16 |
+
step 140000 | 1705s | train_loss=0.1307
|
| 17 |
+
step 150000 | 1815s | train_loss=0.1154
|
| 18 |
+
step 160000 | 1925s | train_loss=0.1167
|
| 19 |
+
step 170000 | 2035s | train_loss=0.1228
|
| 20 |
+
step 180000 | 2145s | train_loss=0.1166
|
| 21 |
+
step 190000 | 2256s | train_loss=0.1029
|
| 22 |
+
step 200000 | 2448s | train_loss=0.1032 | fid=234.8408
|
| 23 |
+
step 210000 | 2559s | train_loss=0.1028
|
| 24 |
+
step 220000 | 2665s | train_loss=0.0908
|
| 25 |
+
step 230000 | 2774s | train_loss=0.0873
|
| 26 |
+
step 240000 | 2884s | train_loss=0.0827
|
| 27 |
+
step 250000 | 2994s | train_loss=0.0854
|
| 28 |
+
step 260000 | 3105s | train_loss=0.0914
|
| 29 |
+
step 270000 | 3215s | train_loss=0.0813
|
| 30 |
+
step 280000 | 3325s | train_loss=0.0835
|
| 31 |
+
step 290000 | 3435s | train_loss=0.0764
|
| 32 |
+
step 300000 | 3626s | train_loss=0.0731 | fid=290.9315
|
| 33 |
+
step 310000 | 3737s | train_loss=0.0785
|
| 34 |
+
step 320000 | 3845s | train_loss=0.0720
|
| 35 |
+
step 330000 | 3953s | train_loss=0.0758
|
| 36 |
+
step 340000 | 4063s | train_loss=0.0689
|
| 37 |
+
step 350000 | 4174s | train_loss=0.0668
|
| 38 |
+
step 360000 | 4284s | train_loss=0.0674
|
| 39 |
+
step 370000 | 4394s | train_loss=0.0658
|
| 40 |
+
step 380000 | 4505s | train_loss=0.0694
|
| 41 |
+
step 390000 | 4615s | train_loss=0.0662
|
| 42 |
+
step 400000 | 4808s | train_loss=0.0615 | fid=325.2368
|
| 43 |
+
step 410000 | 4918s | train_loss=0.0612
|
| 44 |
+
step 420000 | 5028s | train_loss=0.0611
|
| 45 |
+
step 430000 | 5134s | train_loss=0.0570
|
| 46 |
+
step 440000 | 5243s | train_loss=0.0605
|
| 47 |
+
step 450000 | 5354s | train_loss=0.0588
|
| 48 |
+
step 460000 | 5464s | train_loss=0.0566
|
| 49 |
+
step 470000 | 5574s | train_loss=0.0573
|
| 50 |
+
step 480000 | 5684s | train_loss=0.0604
|
| 51 |
+
step 490000 | 5792s | train_loss=0.0559
|
| 52 |
+
step 500000 | 5988s | train_loss=0.0586 | fid=317.5952
|
| 53 |
+
step 510000 | 6098s | train_loss=0.0584
|
| 54 |
+
step 520000 | 6209s | train_loss=0.0568
|
| 55 |
+
step 530000 | 6319s | train_loss=0.0537
|
| 56 |
+
step 540000 | 6425s | train_loss=0.0579
|
| 57 |
+
step 550000 | 6535s | train_loss=0.0549
|
| 58 |
+
step 560000 | 6645s | train_loss=0.0499
|
| 59 |
+
step 570000 | 6755s | train_loss=0.0501
|
| 60 |
+
step 580000 | 6865s | train_loss=0.0509
|
| 61 |
+
step 590000 | 6974s | train_loss=0.0521
|
| 62 |
+
step 600000 | 7169s | train_loss=0.0512 | fid=317.6370
|
| 63 |
+
step 610000 | 7280s | train_loss=0.0542
|
| 64 |
+
step 620000 | 7390s | train_loss=0.0488
|
| 65 |
+
step 630000 | 7500s | train_loss=0.0487
|
| 66 |
+
step 640000 | 7608s | train_loss=0.0504
|
| 67 |
+
step 650000 | 7716s | train_loss=0.0503
|
| 68 |
+
step 660000 | 7826s | train_loss=0.0479
|
| 69 |
+
step 670000 | 7937s | train_loss=0.0481
|
| 70 |
+
step 680000 | 8045s | train_loss=0.0454
|
| 71 |
+
step 690000 | 8155s | train_loss=0.0461
|
| 72 |
+
step 700000 | 8348s | train_loss=0.0480 | fid=296.4231
|
| 73 |
+
step 710000 | 8458s | train_loss=0.0447
|
| 74 |
+
step 720000 | 8568s | train_loss=0.0451
|
| 75 |
+
step 730000 | 8679s | train_loss=0.0458
|
| 76 |
+
step 740000 | 8789s | train_loss=0.0457
|
| 77 |
+
step 750000 | 8897s | train_loss=0.0438
|
| 78 |
+
step 760000 | 9005s | train_loss=0.0450
|
| 79 |
+
step 770000 | 9116s | train_loss=0.0443
|
| 80 |
+
step 780000 | 9224s | train_loss=0.0461
|
| 81 |
+
step 790000 | 9334s | train_loss=0.0427
|
| 82 |
+
step 800000 | 9530s | train_loss=0.0418 | fid=298.6713
|
| 83 |
+
step 810000 | 9640s | train_loss=0.0424
|
| 84 |
+
step 820000 | 9751s | train_loss=0.0438
|
| 85 |
+
step 830000 | 9861s | train_loss=0.0415
|
| 86 |
+
step 840000 | 9971s | train_loss=0.0441
|
| 87 |
+
step 850000 | 10082s | train_loss=0.0428
|
| 88 |
+
step 860000 | 10188s | train_loss=0.0418
|
| 89 |
+
step 870000 | 10298s | train_loss=0.0422
|
| 90 |
+
step 880000 | 10406s | train_loss=0.0394
|
| 91 |
+
step 890000 | 10515s | train_loss=0.0415
|
| 92 |
+
step 900000 | 10712s | train_loss=0.0406 | fid=288.7102
|
| 93 |
+
step 910000 | 10823s | train_loss=0.0429
|
| 94 |
+
step 920000 | 10933s | train_loss=0.0392
|
| 95 |
+
step 930000 | 11043s | train_loss=0.0386
|
| 96 |
+
step 940000 | 11154s | train_loss=0.0443
|
| 97 |
+
step 950000 | 11263s | train_loss=0.0403
|
| 98 |
+
step 960000 | 11372s | train_loss=0.0391
|
| 99 |
+
step 970000 | 11473s | train_loss=0.0409
|
| 100 |
+
step 980000 | 11583s | train_loss=0.0411
|
| 101 |
+
step 990000 | 11692s | train_loss=0.0377
|
| 102 |
+
step 1000000 | 11889s | train_loss=0.0361 | fid=281.9000
|
| 103 |
+
step 1010000 | 12000s | train_loss=0.0380
|
| 104 |
+
step 1020000 | 12110s | train_loss=0.0364
|
| 105 |
+
step 1030000 | 12220s | train_loss=0.0426
|
| 106 |
+
step 1040000 | 12330s | train_loss=0.0383
|
| 107 |
+
step 1050000 | 12441s | train_loss=0.0383
|
| 108 |
+
step 1060000 | 12551s | train_loss=0.0406
|
| 109 |
+
step 1070000 | 12657s | train_loss=0.0357
|
| 110 |
+
step 1080000 | 12765s | train_loss=0.0374
|
| 111 |
+
step 1090000 | 12874s | train_loss=0.0389
|
| 112 |
+
step 1100000 | 13071s | train_loss=0.0363 | fid=268.8231
|
| 113 |
+
step 1110000 | 13181s | train_loss=0.0374
|
| 114 |
+
step 1120000 | 13291s | train_loss=0.0355
|
| 115 |
+
step 1130000 | 13401s | train_loss=0.0459
|
| 116 |
+
step 1140000 | 13511s | train_loss=0.0348
|
| 117 |
+
step 1150000 | 13616s | train_loss=0.0371
|
| 118 |
+
step 1160000 | 13705s | train_loss=0.0369
|
| 119 |
+
step 1170000 | 13795s | train_loss=0.0368
|
| 120 |
+
step 1180000 | 13886s | train_loss=0.0355
|
| 121 |
+
step 1190000 | 13959s | train_loss=0.0346
|
| 122 |
+
step 1200000 | 14155s | train_loss=0.0358 | fid=239.0504
|
| 123 |
+
step 1210000 | 14246s | train_loss=0.0330
|
| 124 |
+
step 1220000 | 14338s | train_loss=0.0366
|
| 125 |
+
step 1230000 | 14429s | train_loss=0.0355
|
| 126 |
+
step 1240000 | 14521s | train_loss=0.0342
|
| 127 |
+
step 1250000 | 14612s | train_loss=0.0375
|
| 128 |
+
step 1260000 | 14701s | train_loss=0.0378
|
| 129 |
+
step 1270000 | 14792s | train_loss=0.0369
|
| 130 |
+
step 1280000 | 14880s | train_loss=0.0339
|
| 131 |
+
step 1290000 | 14972s | train_loss=0.0334
|
| 132 |
+
step 1300000 | 15133s | train_loss=0.0375 | fid=238.1326
|
| 133 |
+
step 1310000 | 15224s | train_loss=0.0318
|
| 134 |
+
step 1320000 | 15316s | train_loss=0.0350
|
| 135 |
+
step 1330000 | 15407s | train_loss=0.0338
|
| 136 |
+
step 1340000 | 15499s | train_loss=0.0350
|
| 137 |
+
step 1350000 | 15588s | train_loss=0.0361
|
| 138 |
+
step 1360000 | 15680s | train_loss=0.0315
|
| 139 |
+
step 1370000 | 15771s | train_loss=0.0309
|
| 140 |
+
step 1380000 | 15859s | train_loss=0.0312
|
| 141 |
+
step 1390000 | 15950s | train_loss=0.0334
|
| 142 |
+
step 1400000 | 16112s | train_loss=0.0351 | fid=227.3125
|
| 143 |
+
step 1410000 | 16204s | train_loss=0.0337
|
| 144 |
+
step 1420000 | 16295s | train_loss=0.0335
|
| 145 |
+
step 1430000 | 16387s | train_loss=0.0321
|
| 146 |
+
step 1440000 | 16478s | train_loss=0.0323
|
| 147 |
+
step 1450000 | 16568s | train_loss=0.0312
|
| 148 |
+
step 1460000 | 16659s | train_loss=0.0329
|
| 149 |
+
step 1470000 | 16751s | train_loss=0.0347
|
| 150 |
+
step 1480000 | 16841s | train_loss=0.0306
|
| 151 |
+
step 1490000 | 16930s | train_loss=0.0328
|
| 152 |
+
step 1500000 | 17090s | train_loss=0.0331 | fid=201.7755
|
| 153 |
+
step 1510000 | 17181s | train_loss=0.0318
|
| 154 |
+
step 1520000 | 17272s | train_loss=0.0395
|
| 155 |
+
step 1530000 | 17364s | train_loss=0.0342
|
| 156 |
+
step 1540000 | 17453s | train_loss=0.0306
|
| 157 |
+
step 1550000 | 17545s | train_loss=0.0357
|
| 158 |
+
step 1560000 | 17636s | train_loss=0.0332
|
| 159 |
+
step 1570000 | 17726s | train_loss=0.0299
|
| 160 |
+
step 1580000 | 17817s | train_loss=0.0300
|
| 161 |
+
step 1590000 | 17907s | train_loss=0.0307
|
| 162 |
+
step 1600000 | 18072s | train_loss=0.0364 | fid=175.0718
|
| 163 |
+
step 1610000 | 18161s | train_loss=0.0338
|
| 164 |
+
step 1620000 | 18252s | train_loss=0.0308
|
| 165 |
+
step 1630000 | 18344s | train_loss=0.0297
|
| 166 |
+
step 1640000 | 18434s | train_loss=0.0332
|
| 167 |
+
step 1650000 | 18525s | train_loss=0.0313
|
| 168 |
+
step 1660000 | 18617s | train_loss=0.0336
|
| 169 |
+
step 1670000 | 18707s | train_loss=0.0319
|
| 170 |
+
step 1680000 | 18798s | train_loss=0.0292
|
| 171 |
+
step 1690000 | 18889s | train_loss=0.0323
|
| 172 |
+
step 1700000 | 19075s | train_loss=0.0292 | fid=176.5134
|
| 173 |
+
step 1710000 | 19161s | train_loss=0.0295
|
| 174 |
+
step 1720000 | 19239s | train_loss=0.0287
|
| 175 |
+
step 1730000 | 19329s | train_loss=0.0310
|
| 176 |
+
step 1740000 | 19420s | train_loss=0.0278
|
| 177 |
+
step 1750000 | 19512s | train_loss=0.0330
|
| 178 |
+
step 1760000 | 19603s | train_loss=0.0319
|
| 179 |
+
step 1770000 | 19693s | train_loss=0.0299
|
| 180 |
+
step 1780000 | 19784s | train_loss=0.0303
|
| 181 |
+
step 1790000 | 19875s | train_loss=0.0310
|
| 182 |
+
step 1800000 | 20077s | train_loss=0.0300 | fid=163.0907
|
| 183 |
+
step 1810000 | 20168s | train_loss=0.0298
|
| 184 |
+
step 1820000 | 20253s | train_loss=0.0309
|
| 185 |
+
step 1830000 | 20338s | train_loss=0.0317
|
| 186 |
+
step 1840000 | 20429s | train_loss=0.0288
|
| 187 |
+
step 1850000 | 20521s | train_loss=0.0300
|
| 188 |
+
step 1860000 | 20610s | train_loss=0.0307
|
| 189 |
+
step 1870000 | 20702s | train_loss=0.0304
|
| 190 |
+
step 1880000 | 20793s | train_loss=0.0290
|
| 191 |
+
step 1890000 | 20884s | train_loss=0.0301
|
| 192 |
+
step 1900000 | 21075s | train_loss=0.0287 | fid=158.5597
|
| 193 |
+
step 1910000 | 21164s | train_loss=0.0286
|
| 194 |
+
step 1920000 | 21255s | train_loss=0.0273
|
| 195 |
+
step 1930000 | 21338s | train_loss=0.0275
|
| 196 |
+
step 1940000 | 21423s | train_loss=0.0302
|
| 197 |
+
step 1950000 | 21513s | train_loss=0.0264
|
| 198 |
+
step 1960000 | 21604s | train_loss=0.0273
|
| 199 |
+
step 1970000 | 21696s | train_loss=0.0286
|
| 200 |
+
step 1980000 | 21787s | train_loss=0.0281
|
| 201 |
+
step 1990000 | 21878s | train_loss=0.0269
|
| 202 |
+
step 2000000 | 22071s | train_loss=0.0266 | fid=142.9785
|
| 203 |
+
step 2010000 | 22160s | train_loss=0.0317
|
| 204 |
+
step 2020000 | 22252s | train_loss=0.0315
|
| 205 |
+
step 2030000 | 22343s | train_loss=0.0287
|
| 206 |
+
step 2040000 | 22424s | train_loss=0.0274
|
| 207 |
+
step 2050000 | 22509s | train_loss=0.0285
|
| 208 |
+
step 2060000 | 22601s | train_loss=0.0312
|
| 209 |
+
step 2070000 | 22692s | train_loss=0.0280
|
| 210 |
+
step 2080000 | 22783s | train_loss=0.0287
|
| 211 |
+
step 2090000 | 22874s | train_loss=0.0274
|
| 212 |
+
step 2100000 | 23069s | train_loss=0.0295 | fid=136.4118
|
| 213 |
+
step 2110000 | 23160s | train_loss=0.0272
|
| 214 |
+
step 2120000 | 23252s | train_loss=0.0278
|
| 215 |
+
step 2130000 | 23343s | train_loss=0.0288
|
| 216 |
+
step 2140000 | 23433s | train_loss=0.0296
|
| 217 |
+
step 2150000 | 23512s | train_loss=0.0256
|
| 218 |
+
step 2160000 | 23601s | train_loss=0.0278
|
| 219 |
+
step 2170000 | 23692s | train_loss=0.0284
|
| 220 |
+
step 2180000 | 23784s | train_loss=0.0258
|
| 221 |
+
step 2190000 | 23875s | train_loss=0.0250
|
| 222 |
+
step 2200000 | 24085s | train_loss=0.0253 | fid=129.1658
|
| 223 |
+
step 2210000 | 24176s | train_loss=0.0258
|
| 224 |
+
step 2220000 | 24268s | train_loss=0.0295
|
| 225 |
+
step 2230000 | 24357s | train_loss=0.0271
|
| 226 |
+
step 2240000 | 24449s | train_loss=0.0246
|
| 227 |
+
step 2250000 | 24541s | train_loss=0.0298
|
| 228 |
+
step 2260000 | 24622s | train_loss=0.0260
|
| 229 |
+
step 2270000 | 24713s | train_loss=0.0294
|
| 230 |
+
step 2280000 | 24804s | train_loss=0.0300
|
| 231 |
+
step 2290000 | 24894s | train_loss=0.0299
|
| 232 |
+
step 2300000 | 25081s | train_loss=0.0259 | fid=133.4185
|
| 233 |
+
step 2310000 | 25172s | train_loss=0.0283
|
| 234 |
+
step 2320000 | 25263s | train_loss=0.0268
|
| 235 |
+
step 2330000 | 25353s | train_loss=0.0267
|
| 236 |
+
step 2340000 | 25444s | train_loss=0.0286
|
| 237 |
+
step 2350000 | 25536s | train_loss=0.0262
|
| 238 |
+
step 2360000 | 25623s | train_loss=0.0306
|
| 239 |
+
step 2370000 | 25708s | train_loss=0.0277
|
| 240 |
+
step 2380000 | 25798s | train_loss=0.0255
|
| 241 |
+
step 2390000 | 25889s | train_loss=0.0272
|
| 242 |
+
step 2400000 | 26095s | train_loss=0.0269 | fid=137.3254
|
| 243 |
+
step 2410000 | 26187s | train_loss=0.0237
|
| 244 |
+
step 2420000 | 26276s | train_loss=0.0259
|
| 245 |
+
step 2430000 | 26367s | train_loss=0.0264
|
| 246 |
+
step 2440000 | 26459s | train_loss=0.0247
|
| 247 |
+
step 2450000 | 26550s | train_loss=0.0301
|
| 248 |
+
step 2460000 | 26641s | train_loss=0.0271
|
| 249 |
+
step 2470000 | 26716s | train_loss=0.0295
|
| 250 |
+
step 2480000 | 26805s | train_loss=0.0242
|
| 251 |
+
step 2490000 | 26897s | train_loss=0.0254
|
| 252 |
+
step 2500000 | 27094s | train_loss=0.0245 | fid=120.5062
|
| 253 |
+
step 2510000 | 27184s | train_loss=0.0233
|
| 254 |
+
step 2520000 | 27275s | train_loss=0.0272
|
| 255 |
+
step 2530000 | 27367s | train_loss=0.0255
|
| 256 |
+
step 2540000 | 27458s | train_loss=0.0250
|
| 257 |
+
step 2550000 | 27549s | train_loss=0.0265
|
| 258 |
+
step 2560000 | 27638s | train_loss=0.0252
|
| 259 |
+
step 2570000 | 27730s | train_loss=0.0251
|
| 260 |
+
step 2580000 | 27806s | train_loss=0.0248
|
| 261 |
+
step 2590000 | 27896s | train_loss=0.0246
|
| 262 |
+
step 2600000 | 28101s | train_loss=0.0281 | fid=126.1381
|
| 263 |
+
step 2610000 | 28192s | train_loss=0.0264
|
| 264 |
+
step 2620000 | 28283s | train_loss=0.0241
|
| 265 |
+
step 2630000 | 28375s | train_loss=0.0254
|
| 266 |
+
step 2640000 | 28466s | train_loss=0.0272
|
| 267 |
+
step 2650000 | 28555s | train_loss=0.0266
|
| 268 |
+
step 2660000 | 28646s | train_loss=0.0237
|
| 269 |
+
step 2670000 | 28738s | train_loss=0.0240
|
| 270 |
+
step 2680000 | 28830s | train_loss=0.0238
|
| 271 |
+
step 2690000 | 28910s | train_loss=0.0240
|
| 272 |
+
step 2700000 | 29113s | train_loss=0.0250 | fid=122.3632
|
| 273 |
+
step 2710000 | 29205s | train_loss=0.0299
|
| 274 |
+
step 2720000 | 29296s | train_loss=0.0260
|
| 275 |
+
step 2730000 | 29387s | train_loss=0.0219
|
| 276 |
+
step 2740000 | 29477s | train_loss=0.0288
|
| 277 |
+
step 2750000 | 29568s | train_loss=0.0228
|
| 278 |
+
step 2760000 | 29660s | train_loss=0.0226
|
| 279 |
+
step 2770000 | 29751s | train_loss=0.0221
|
| 280 |
+
step 2780000 | 29843s | train_loss=0.0228
|
| 281 |
+
step 2790000 | 29927s | train_loss=0.0239
|
| 282 |
+
step 2800000 | 30132s | train_loss=0.0234 | fid=122.2715
|
| 283 |
+
step 2810000 | 30224s | train_loss=0.0246
|
| 284 |
+
step 2820000 | 30315s | train_loss=0.0244
|
| 285 |
+
step 2830000 | 30404s | train_loss=0.0254
|
| 286 |
+
step 2840000 | 30496s | train_loss=0.0233
|
| 287 |
+
step 2850000 | 30587s | train_loss=0.0239
|
| 288 |
+
step 2860000 | 30678s | train_loss=0.0221
|
| 289 |
+
step 2870000 | 30770s | train_loss=0.0251
|
| 290 |
+
step 2880000 | 30861s | train_loss=0.0236
|
| 291 |
+
step 2890000 | 30951s | train_loss=0.0265
|
| 292 |
+
step 2900000 | 31132s | train_loss=0.0236 | fid=122.6633
|
| 293 |
+
step 2910000 | 31224s | train_loss=0.0234
|
| 294 |
+
step 2920000 | 31313s | train_loss=0.0250
|
| 295 |
+
step 2930000 | 31405s | train_loss=0.0227
|
| 296 |
+
step 2940000 | 31496s | train_loss=0.0227
|
| 297 |
+
step 2950000 | 31587s | train_loss=0.0257
|
| 298 |
+
step 2960000 | 31679s | train_loss=0.0226
|
| 299 |
+
step 2970000 | 31770s | train_loss=0.0228
|
| 300 |
+
step 2980000 | 31861s | train_loss=0.0225
|
| 301 |
+
step 2990000 | 31951s | train_loss=0.0247
|
| 302 |
+
step 3000000 | 32146s | train_loss=0.0245 | fid=121.0720
|
| 303 |
+
step 3010000 | 32235s | train_loss=0.0250
|
| 304 |
+
step 3020000 | 32326s | train_loss=0.0233
|
| 305 |
+
step 3030000 | 32417s | train_loss=0.0308
|
| 306 |
+
step 3040000 | 32509s | train_loss=0.0237
|
| 307 |
+
step 3050000 | 32600s | train_loss=0.0229
|
| 308 |
+
step 3060000 | 32692s | train_loss=0.0247
|
| 309 |
+
step 3070000 | 32783s | train_loss=0.0222
|
| 310 |
+
step 3080000 | 32873s | train_loss=0.0246
|
| 311 |
+
step 3090000 | 32964s | train_loss=0.0255
|
| 312 |
+
step 3100000 | 33136s | train_loss=0.0267 | fid=122.3822
|
| 313 |
+
step 3110000 | 33225s | train_loss=0.0247
|
| 314 |
+
step 3120000 | 33317s | train_loss=0.0240
|
| 315 |
+
step 3130000 | 33408s | train_loss=0.0251
|
| 316 |
+
step 3140000 | 33499s | train_loss=0.0274
|
| 317 |
+
step 3150000 | 33590s | train_loss=0.0222
|
| 318 |
+
step 3160000 | 33681s | train_loss=0.0240
|
| 319 |
+
step 3170000 | 33771s | train_loss=0.0256
|
| 320 |
+
step 3180000 | 33862s | train_loss=0.0237
|
| 321 |
+
step 3190000 | 33953s | train_loss=0.0236
|
| 322 |
+
step 3200000 | 34139s | train_loss=0.0227 | fid=120.7686
|
| 323 |
+
step 3210000 | 34217s | train_loss=0.0248
|
| 324 |
+
step 3220000 | 34309s | train_loss=0.0223
|
| 325 |
+
step 3230000 | 34400s | train_loss=0.0232
|
| 326 |
+
step 3240000 | 34491s | train_loss=0.0226
|
| 327 |
+
step 3250000 | 34582s | train_loss=0.0262
|
| 328 |
+
step 3260000 | 34674s | train_loss=0.0231
|
| 329 |
+
step 3270000 | 34763s | train_loss=0.0254
|
| 330 |
+
step 3280000 | 34855s | train_loss=0.0255
|
| 331 |
+
step 3290000 | 34946s | train_loss=0.0222
|
| 332 |
+
step 3300000 | 35156s | train_loss=0.0231 | fid=122.2512
|
| 333 |
+
step 3310000 | 35243s | train_loss=0.0210
|
| 334 |
+
step 3320000 | 35328s | train_loss=0.0237
|
| 335 |
+
step 3330000 | 35419s | train_loss=0.0213
|
| 336 |
+
step 3340000 | 35511s | train_loss=0.0259
|
| 337 |
+
step 3350000 | 35602s | train_loss=0.0245
|
| 338 |
+
step 3360000 | 35691s | train_loss=0.0212
|
| 339 |
+
step 3370000 | 35783s | train_loss=0.0265
|
| 340 |
+
step 3380000 | 35874s | train_loss=0.0256
|
| 341 |
+
step 3390000 | 35964s | train_loss=0.0252
|
| 342 |
+
step 3400000 | 36155s | train_loss=0.0230 | fid=116.6327
|
| 343 |
+
step 3410000 | 36246s | train_loss=0.0249
|
| 344 |
+
step 3420000 | 36328s | train_loss=0.0235
|
| 345 |
+
step 3430000 | 36420s | train_loss=0.0227
|
| 346 |
+
step 3440000 | 36511s | train_loss=0.0237
|
| 347 |
+
step 3450000 | 36601s | train_loss=0.0224
|
| 348 |
+
step 3460000 | 36692s | train_loss=0.0204
|
| 349 |
+
step 3470000 | 36784s | train_loss=0.0215
|
| 350 |
+
step 3480000 | 36873s | train_loss=0.0202
|
| 351 |
+
step 3490000 | 36964s | train_loss=0.0215
|
| 352 |
+
step 3500000 | 37158s | train_loss=0.0221 | fid=140.0636
|
| 353 |
+
step 3510000 | 37250s | train_loss=0.0222
|
| 354 |
+
step 3520000 | 37341s | train_loss=0.0221
|
| 355 |
+
step 3530000 | 37418s | train_loss=0.0228
|
| 356 |
+
step 3540000 | 37509s | train_loss=0.0198
|
| 357 |
+
step 3550000 | 37599s | train_loss=0.0218
|
| 358 |
+
step 3560000 | 37690s | train_loss=0.0212
|
| 359 |
+
step 3570000 | 37779s | train_loss=0.0227
|
| 360 |
+
step 3580000 | 37871s | train_loss=0.0238
|
| 361 |
+
step 3590000 | 37962s | train_loss=0.0227
|
| 362 |
+
step 3600000 | 38177s | train_loss=0.0219 | fid=130.5959
|
| 363 |
+
step 3610000 | 38268s | train_loss=0.0229
|
| 364 |
+
step 3620000 | 38359s | train_loss=0.0211
|
| 365 |
+
step 3630000 | 38449s | train_loss=0.0233
|
| 366 |
+
step 3640000 | 38538s | train_loss=0.0237
|
| 367 |
+
step 3650000 | 38629s | train_loss=0.0282
|
| 368 |
+
step 3660000 | 38719s | train_loss=0.0221
|
| 369 |
+
step 3670000 | 38810s | train_loss=0.0278
|
| 370 |
+
step 3680000 | 38901s | train_loss=0.0218
|
| 371 |
+
step 3690000 | 38993s | train_loss=0.0219
|
| 372 |
+
step 3700000 | 39158s | train_loss=0.0223 | fid=134.4930
|
| 373 |
+
step 3710000 | 39248s | train_loss=0.0216
|
| 374 |
+
step 3720000 | 39340s | train_loss=0.0250
|
| 375 |
+
step 3730000 | 39431s | train_loss=0.0221
|
| 376 |
+
step 3740000 | 39519s | train_loss=0.0223
|
| 377 |
+
step 3750000 | 39610s | train_loss=0.0227
|
| 378 |
+
step 3760000 | 39700s | train_loss=0.0212
|
| 379 |
+
step 3770000 | 39791s | train_loss=0.0246
|
| 380 |
+
step 3780000 | 39883s | train_loss=0.0215
|
| 381 |
+
step 3790000 | 39974s | train_loss=0.0224
|
| 382 |
+
step 3800000 | 40139s | train_loss=0.0207 | fid=124.1664
|
| 383 |
+
step 3810000 | 40229s | train_loss=0.0237
|
| 384 |
+
step 3820000 | 40321s | train_loss=0.0208
|
| 385 |
+
step 3830000 | 40410s | train_loss=0.0216
|
| 386 |
+
step 3840000 | 40502s | train_loss=0.0216
|
| 387 |
+
step 3850000 | 40577s | train_loss=0.0237
|
| 388 |
+
step 3860000 | 40668s | train_loss=0.0214
|
| 389 |
+
step 3870000 | 40759s | train_loss=0.0217
|
| 390 |
+
step 3880000 | 40851s | train_loss=0.0239
|
| 391 |
+
step 3890000 | 40942s | train_loss=0.0221
|
| 392 |
+
step 3900000 | 41153s | train_loss=0.0211 | fid=130.6428
|
| 393 |
+
step 3910000 | 41243s | train_loss=0.0224
|
| 394 |
+
step 3920000 | 41332s | train_loss=0.0212
|
| 395 |
+
step 3930000 | 41424s | train_loss=0.0229
|
| 396 |
+
step 3940000 | 41513s | train_loss=0.0222
|
| 397 |
+
step 3950000 | 41599s | train_loss=0.0216
|
| 398 |
+
step 3960000 | 41682s | train_loss=0.0223
|
| 399 |
+
step 3970000 | 41773s | train_loss=0.0238
|
| 400 |
+
step 3980000 | 41865s | train_loss=0.0194
|
| 401 |
+
step 3990000 | 41956s | train_loss=0.0258
|
| 402 |
+
step 4000000 | 42175s | train_loss=0.0205 | fid=132.2727
|
| 403 |
+
step 4010000 | 42264s | train_loss=0.0209
|
| 404 |
+
step 4020000 | 42354s | train_loss=0.0213
|
| 405 |
+
step 4030000 | 42444s | train_loss=0.0226
|
| 406 |
+
step 4040000 | 42535s | train_loss=0.0223
|
| 407 |
+
step 4050000 | 42626s | train_loss=0.0202
|
| 408 |
+
step 4060000 | 42703s | train_loss=0.0203
|
| 409 |
+
step 4070000 | 42790s | train_loss=0.0196
|
| 410 |
+
step 4080000 | 42881s | train_loss=0.0235
|
| 411 |
+
step 4090000 | 42972s | train_loss=0.0234
|
| 412 |
+
step 4100000 | 43187s | train_loss=0.0233 | fid=137.5763
|
| 413 |
+
step 4110000 | 43275s | train_loss=0.0201
|
| 414 |
+
step 4120000 | 43365s | train_loss=0.0234
|
| 415 |
+
step 4130000 | 43456s | train_loss=0.0206
|
| 416 |
+
step 4140000 | 43548s | train_loss=0.0236
|
| 417 |
+
step 4150000 | 43639s | train_loss=0.0194
|
| 418 |
+
step 4160000 | 43731s | train_loss=0.0208
|
| 419 |
+
step 4170000 | 43809s | train_loss=0.0208
|
| 420 |
+
step 4180000 | 43900s | train_loss=0.0261
|
| 421 |
+
step 4190000 | 43992s | train_loss=0.0228
|
| 422 |
+
step 4200000 | 44208s | train_loss=0.0208 | fid=139.8566
|
| 423 |
+
step 4210000 | 44295s | train_loss=0.0217
|
| 424 |
+
step 4220000 | 44387s | train_loss=0.0235
|
| 425 |
+
step 4230000 | 44478s | train_loss=0.0210
|
| 426 |
+
step 4240000 | 44569s | train_loss=0.0209
|
| 427 |
+
step 4250000 | 44661s | train_loss=0.0235
|
| 428 |
+
step 4260000 | 44752s | train_loss=0.0206
|
| 429 |
+
step 4270000 | 44837s | train_loss=0.0194
|
| 430 |
+
step 4280000 | 44922s | train_loss=0.0205
|
| 431 |
+
step 4290000 | 45014s | train_loss=0.0212
|
| 432 |
+
step 4300000 | 45222s | train_loss=0.0206 | fid=132.6253
|
| 433 |
+
step 4310000 | 45312s | train_loss=0.0197
|
| 434 |
+
step 4320000 | 45403s | train_loss=0.0193
|
| 435 |
+
step 4330000 | 45495s | train_loss=0.0196
|
| 436 |
+
step 4340000 | 45586s | train_loss=0.0244
|
| 437 |
+
step 4350000 | 45677s | train_loss=0.0224
|
| 438 |
+
step 4360000 | 45769s | train_loss=0.0200
|
| 439 |
+
step 4370000 | 45860s | train_loss=0.0238
|
| 440 |
+
step 4380000 | 45936s | train_loss=0.0209
|
| 441 |
+
step 4390000 | 46021s | train_loss=0.0222
|
| 442 |
+
step 4400000 | 46232s | train_loss=0.0213 | fid=134.0848
|
| 443 |
+
step 4410000 | 46322s | train_loss=0.0225
|
| 444 |
+
step 4420000 | 46414s | train_loss=0.0218
|
| 445 |
+
step 4430000 | 46505s | train_loss=0.0219
|
| 446 |
+
step 4440000 | 46596s | train_loss=0.0189
|
| 447 |
+
step 4450000 | 46687s | train_loss=0.0212
|
| 448 |
+
step 4460000 | 46779s | train_loss=0.0215
|
| 449 |
+
step 4470000 | 46870s | train_loss=0.0188
|
| 450 |
+
step 4480000 | 46961s | train_loss=0.0214
|
| 451 |
+
step 4490000 | 47034s | train_loss=0.0225
|
| 452 |
+
step 4500000 | 47237s | train_loss=0.0220 | fid=123.2409
|
| 453 |
+
step 4510000 | 47327s | train_loss=0.0209
|
| 454 |
+
step 4520000 | 47418s | train_loss=0.0254
|
| 455 |
+
step 4530000 | 47509s | train_loss=0.0190
|
| 456 |
+
step 4540000 | 47601s | train_loss=0.0200
|
| 457 |
+
step 4550000 | 47692s | train_loss=0.0298
|
| 458 |
+
step 4560000 | 47784s | train_loss=0.0204
|
| 459 |
+
step 4570000 | 47875s | train_loss=0.0191
|
| 460 |
+
step 4580000 | 47961s | train_loss=0.0203
|
| 461 |
+
step 4590000 | 48052s | train_loss=0.0205
|
| 462 |
+
step 4600000 | 48216s | train_loss=0.0212 | fid=128.4838
|
| 463 |
+
step 4610000 | 48306s | train_loss=0.0196
|
| 464 |
+
step 4620000 | 48397s | train_loss=0.0226
|
| 465 |
+
step 4630000 | 48488s | train_loss=0.0224
|
| 466 |
+
step 4640000 | 48580s | train_loss=0.0203
|
| 467 |
+
step 4650000 | 48671s | train_loss=0.0191
|
| 468 |
+
step 4660000 | 48763s | train_loss=0.0209
|
| 469 |
+
step 4670000 | 48854s | train_loss=0.0199
|
| 470 |
+
step 4680000 | 48930s | train_loss=0.0208
|
| 471 |
+
step 4690000 | 49004s | train_loss=0.0200
|
| 472 |
+
step 4700000 | 49136s | train_loss=0.0228 | fid=131.8201
|
| 473 |
+
step 4710000 | 49210s | train_loss=0.0200
|
| 474 |
+
step 4720000 | 49282s | train_loss=0.0181
|
| 475 |
+
step 4730000 | 49355s | train_loss=0.0195
|
| 476 |
+
step 4740000 | 49429s | train_loss=0.0214
|
| 477 |
+
step 4750000 | 49503s | train_loss=0.0190
|
| 478 |
+
step 4760000 | 49577s | train_loss=0.0186
|
| 479 |
+
step 4770000 | 49647s | train_loss=0.0245
|
| 480 |
+
step 4780000 | 49703s | train_loss=0.0199
|
| 481 |
+
step 4790000 | 49759s | train_loss=0.0208
|
| 482 |
+
step 4800000 | 49858s | train_loss=0.0205 | fid=140.9861
|
| 483 |
+
step 4810000 | 49914s | train_loss=0.0198
|
| 484 |
+
step 4820000 | 49971s | train_loss=0.0190
|
| 485 |
+
step 4830000 | 50023s | train_loss=0.0240
|
| 486 |
+
step 4840000 | 50079s | train_loss=0.0206
|
| 487 |
+
step 4850000 | 50135s | train_loss=0.0189
|
| 488 |
+
step 4860000 | 50192s | train_loss=0.0211
|
| 489 |
+
step 4870000 | 50248s | train_loss=0.0177
|
| 490 |
+
step 4880000 | 50304s | train_loss=0.0203
|
| 491 |
+
step 4890000 | 50360s | train_loss=0.0190
|
| 492 |
+
step 4900000 | 50462s | train_loss=0.0201 | fid=132.7694
|
| 493 |
+
step 4910000 | 50514s | train_loss=0.0187
|
| 494 |
+
step 4920000 | 50570s | train_loss=0.0195
|
| 495 |
+
step 4930000 | 50623s | train_loss=0.0201
|
| 496 |
+
step 4940000 | 50679s | train_loss=0.0201
|
| 497 |
+
step 4950000 | 50735s | train_loss=0.0220
|
| 498 |
+
step 4960000 | 50791s | train_loss=0.0198
|
| 499 |
+
step 4970000 | 50848s | train_loss=0.0195
|
| 500 |
+
step 4980000 | 50904s | train_loss=0.0190
|
| 501 |
+
step 4990000 | 50960s | train_loss=0.0193
|
| 502 |
+
step 5000000 | 51063s | train_loss=0.0199 | fid=129.9638
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d220/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d220/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d220",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 44,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d220/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d220/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 49s | train_loss=1.0035 | fid=298.7339
|
| 3 |
+
step 10000 | 107s | train_loss=0.7078
|
| 4 |
+
step 20000 | 166s | train_loss=0.5945
|
| 5 |
+
step 30000 | 226s | train_loss=0.5282
|
| 6 |
+
step 40000 | 285s | train_loss=0.4629
|
| 7 |
+
step 50000 | 344s | train_loss=0.3891
|
| 8 |
+
step 60000 | 403s | train_loss=0.3124
|
| 9 |
+
step 70000 | 463s | train_loss=0.2628
|
| 10 |
+
step 80000 | 522s | train_loss=0.2267
|
| 11 |
+
step 90000 | 581s | train_loss=0.2063
|
| 12 |
+
step 100000 | 686s | train_loss=0.1756 | fid=199.4007
|
| 13 |
+
step 110000 | 745s | train_loss=0.1557
|
| 14 |
+
step 120000 | 804s | train_loss=0.1495
|
| 15 |
+
step 130000 | 863s | train_loss=0.1552
|
| 16 |
+
step 140000 | 922s | train_loss=0.1340
|
| 17 |
+
step 150000 | 982s | train_loss=0.1260
|
| 18 |
+
step 160000 | 1041s | train_loss=0.1236
|
| 19 |
+
step 170000 | 1101s | train_loss=0.1243
|
| 20 |
+
step 180000 | 1160s | train_loss=0.1222
|
| 21 |
+
step 190000 | 1220s | train_loss=0.1116
|
| 22 |
+
step 200000 | 1324s | train_loss=0.1075 | fid=234.3538
|
| 23 |
+
step 210000 | 1383s | train_loss=0.1134
|
| 24 |
+
step 220000 | 1443s | train_loss=0.1104
|
| 25 |
+
step 230000 | 1502s | train_loss=0.0984
|
| 26 |
+
step 240000 | 1561s | train_loss=0.0928
|
| 27 |
+
step 250000 | 1621s | train_loss=0.0929
|
| 28 |
+
step 260000 | 1680s | train_loss=0.0937
|
| 29 |
+
step 270000 | 1740s | train_loss=0.0873
|
| 30 |
+
step 280000 | 1800s | train_loss=0.0962
|
| 31 |
+
step 290000 | 1857s | train_loss=0.0929
|
| 32 |
+
step 300000 | 1963s | train_loss=0.0893 | fid=272.9030
|
| 33 |
+
step 310000 | 2022s | train_loss=0.0843
|
| 34 |
+
step 320000 | 2081s | train_loss=0.0795
|
| 35 |
+
step 330000 | 2140s | train_loss=0.0846
|
| 36 |
+
step 340000 | 2199s | train_loss=0.0851
|
| 37 |
+
step 350000 | 2258s | train_loss=0.0720
|
| 38 |
+
step 360000 | 2318s | train_loss=0.0793
|
| 39 |
+
step 370000 | 2378s | train_loss=0.0804
|
| 40 |
+
step 380000 | 2437s | train_loss=0.0761
|
| 41 |
+
step 390000 | 2493s | train_loss=0.0697
|
| 42 |
+
step 400000 | 2601s | train_loss=0.0733 | fid=296.2235
|
| 43 |
+
step 410000 | 2661s | train_loss=0.0683
|
| 44 |
+
step 420000 | 2720s | train_loss=0.0657
|
| 45 |
+
step 430000 | 2779s | train_loss=0.0643
|
| 46 |
+
step 440000 | 2838s | train_loss=0.0637
|
| 47 |
+
step 450000 | 2897s | train_loss=0.0636
|
| 48 |
+
step 460000 | 2957s | train_loss=0.0689
|
| 49 |
+
step 470000 | 3017s | train_loss=0.0629
|
| 50 |
+
step 480000 | 3072s | train_loss=0.0682
|
| 51 |
+
step 490000 | 3131s | train_loss=0.0596
|
| 52 |
+
step 500000 | 3239s | train_loss=0.0611 | fid=302.1215
|
| 53 |
+
step 510000 | 3298s | train_loss=0.0634
|
| 54 |
+
step 520000 | 3358s | train_loss=0.0598
|
| 55 |
+
step 530000 | 3417s | train_loss=0.0623
|
| 56 |
+
step 540000 | 3477s | train_loss=0.0619
|
| 57 |
+
step 550000 | 3537s | train_loss=0.0680
|
| 58 |
+
step 560000 | 3596s | train_loss=0.0560
|
| 59 |
+
step 570000 | 3656s | train_loss=0.0569
|
| 60 |
+
step 580000 | 3710s | train_loss=0.0610
|
| 61 |
+
step 590000 | 3770s | train_loss=0.0529
|
| 62 |
+
step 600000 | 3879s | train_loss=0.0520 | fid=290.5151
|
| 63 |
+
step 610000 | 3939s | train_loss=0.0568
|
| 64 |
+
step 620000 | 3998s | train_loss=0.0521
|
| 65 |
+
step 630000 | 4057s | train_loss=0.0521
|
| 66 |
+
step 640000 | 4117s | train_loss=0.0545
|
| 67 |
+
step 650000 | 4177s | train_loss=0.0574
|
| 68 |
+
step 660000 | 4236s | train_loss=0.0547
|
| 69 |
+
step 670000 | 4291s | train_loss=0.0516
|
| 70 |
+
step 680000 | 4350s | train_loss=0.0486
|
| 71 |
+
step 690000 | 4409s | train_loss=0.0487
|
| 72 |
+
step 700000 | 4518s | train_loss=0.0486 | fid=287.9598
|
| 73 |
+
step 710000 | 4578s | train_loss=0.0485
|
| 74 |
+
step 720000 | 4637s | train_loss=0.0521
|
| 75 |
+
step 730000 | 4697s | train_loss=0.0477
|
| 76 |
+
step 740000 | 4756s | train_loss=0.0442
|
| 77 |
+
step 750000 | 4815s | train_loss=0.0472
|
| 78 |
+
step 760000 | 4874s | train_loss=0.0461
|
| 79 |
+
step 770000 | 4929s | train_loss=0.0456
|
| 80 |
+
step 780000 | 4988s | train_loss=0.0463
|
| 81 |
+
step 790000 | 5048s | train_loss=0.0470
|
| 82 |
+
step 800000 | 5156s | train_loss=0.0441 | fid=284.0011
|
| 83 |
+
step 810000 | 5215s | train_loss=0.0506
|
| 84 |
+
step 820000 | 5275s | train_loss=0.0489
|
| 85 |
+
step 830000 | 5335s | train_loss=0.0456
|
| 86 |
+
step 840000 | 5394s | train_loss=0.0443
|
| 87 |
+
step 850000 | 5454s | train_loss=0.0442
|
| 88 |
+
step 860000 | 5508s | train_loss=0.0445
|
| 89 |
+
step 870000 | 5564s | train_loss=0.0436
|
| 90 |
+
step 880000 | 5603s | train_loss=0.0446
|
| 91 |
+
step 890000 | 5643s | train_loss=0.0437
|
| 92 |
+
step 900000 | 5718s | train_loss=0.0417 | fid=281.9232
|
| 93 |
+
step 910000 | 5757s | train_loss=0.0422
|
| 94 |
+
step 920000 | 5797s | train_loss=0.0417
|
| 95 |
+
step 930000 | 5836s | train_loss=0.0404
|
| 96 |
+
step 940000 | 5876s | train_loss=0.0490
|
| 97 |
+
step 950000 | 5908s | train_loss=0.0412
|
| 98 |
+
step 960000 | 5942s | train_loss=0.0467
|
| 99 |
+
step 970000 | 5981s | train_loss=0.0425
|
| 100 |
+
step 980000 | 6020s | train_loss=0.0437
|
| 101 |
+
step 990000 | 6060s | train_loss=0.0416
|
| 102 |
+
step 1000000 | 6135s | train_loss=0.0409 | fid=274.9382
|
| 103 |
+
step 1010000 | 6174s | train_loss=0.0417
|
| 104 |
+
step 1020000 | 6213s | train_loss=0.0419
|
| 105 |
+
step 1030000 | 6252s | train_loss=0.0631
|
| 106 |
+
step 1040000 | 6290s | train_loss=0.0401
|
| 107 |
+
step 1050000 | 6323s | train_loss=0.0406
|
| 108 |
+
step 1060000 | 6362s | train_loss=0.0448
|
| 109 |
+
step 1070000 | 6401s | train_loss=0.0434
|
| 110 |
+
step 1080000 | 6440s | train_loss=0.0434
|
| 111 |
+
step 1090000 | 6479s | train_loss=0.0398
|
| 112 |
+
step 1100000 | 6554s | train_loss=0.0366 | fid=264.6617
|
| 113 |
+
step 1110000 | 6594s | train_loss=0.0453
|
| 114 |
+
step 1120000 | 6633s | train_loss=0.0382
|
| 115 |
+
step 1130000 | 6672s | train_loss=0.0459
|
| 116 |
+
step 1140000 | 6704s | train_loss=0.0368
|
| 117 |
+
step 1150000 | 6739s | train_loss=0.0393
|
| 118 |
+
step 1160000 | 6778s | train_loss=0.0407
|
| 119 |
+
step 1170000 | 6817s | train_loss=0.0391
|
| 120 |
+
step 1180000 | 6856s | train_loss=0.0387
|
| 121 |
+
step 1190000 | 6896s | train_loss=0.0388
|
| 122 |
+
step 1200000 | 6970s | train_loss=0.0402 | fid=254.3679
|
| 123 |
+
step 1210000 | 7009s | train_loss=0.0380
|
| 124 |
+
step 1220000 | 7049s | train_loss=0.0448
|
| 125 |
+
step 1230000 | 7087s | train_loss=0.0398
|
| 126 |
+
step 1240000 | 7119s | train_loss=0.0366
|
| 127 |
+
step 1250000 | 7158s | train_loss=0.0450
|
| 128 |
+
step 1260000 | 7198s | train_loss=0.0369
|
| 129 |
+
step 1270000 | 7237s | train_loss=0.0361
|
| 130 |
+
step 1280000 | 7276s | train_loss=0.0400
|
| 131 |
+
step 1290000 | 7316s | train_loss=0.0361
|
| 132 |
+
step 1300000 | 7391s | train_loss=0.0440 | fid=262.5761
|
| 133 |
+
step 1310000 | 7431s | train_loss=0.0358
|
| 134 |
+
step 1320000 | 7470s | train_loss=0.0380
|
| 135 |
+
step 1330000 | 7501s | train_loss=0.0381
|
| 136 |
+
step 1340000 | 7539s | train_loss=0.0370
|
| 137 |
+
step 1350000 | 7578s | train_loss=0.0369
|
| 138 |
+
step 1360000 | 7618s | train_loss=0.0371
|
| 139 |
+
step 1370000 | 7657s | train_loss=0.0325
|
| 140 |
+
step 1380000 | 7697s | train_loss=0.0343
|
| 141 |
+
step 1390000 | 7737s | train_loss=0.0380
|
| 142 |
+
step 1400000 | 7811s | train_loss=0.0363 | fid=262.4124
|
| 143 |
+
step 1410000 | 7850s | train_loss=0.0369
|
| 144 |
+
step 1420000 | 7889s | train_loss=0.0459
|
| 145 |
+
step 1430000 | 7921s | train_loss=0.0343
|
| 146 |
+
step 1440000 | 7960s | train_loss=0.0351
|
| 147 |
+
step 1450000 | 7999s | train_loss=0.0347
|
| 148 |
+
step 1460000 | 8038s | train_loss=0.0373
|
| 149 |
+
step 1470000 | 8077s | train_loss=0.0357
|
| 150 |
+
step 1480000 | 8117s | train_loss=0.0351
|
| 151 |
+
step 1490000 | 8156s | train_loss=0.0345
|
| 152 |
+
step 1500000 | 8230s | train_loss=0.0348 | fid=250.9980
|
| 153 |
+
step 1510000 | 8269s | train_loss=0.0347
|
| 154 |
+
step 1520000 | 8297s | train_loss=0.0407
|
| 155 |
+
step 1530000 | 8333s | train_loss=0.0378
|
| 156 |
+
step 1540000 | 8373s | train_loss=0.0361
|
| 157 |
+
step 1550000 | 8413s | train_loss=0.0334
|
| 158 |
+
step 1560000 | 8452s | train_loss=0.0403
|
| 159 |
+
step 1570000 | 8491s | train_loss=0.0343
|
| 160 |
+
step 1580000 | 8530s | train_loss=0.0344
|
| 161 |
+
step 1590000 | 8570s | train_loss=0.0321
|
| 162 |
+
step 1600000 | 8645s | train_loss=0.0377 | fid=254.2971
|
| 163 |
+
step 1610000 | 8683s | train_loss=0.0394
|
| 164 |
+
step 1620000 | 8713s | train_loss=0.0311
|
| 165 |
+
step 1630000 | 8752s | train_loss=0.0316
|
| 166 |
+
step 1640000 | 8792s | train_loss=0.0362
|
| 167 |
+
step 1650000 | 8831s | train_loss=0.0355
|
| 168 |
+
step 1660000 | 8870s | train_loss=0.0399
|
| 169 |
+
step 1670000 | 8910s | train_loss=0.0368
|
| 170 |
+
step 1680000 | 8949s | train_loss=0.0343
|
| 171 |
+
step 1690000 | 8988s | train_loss=0.0475
|
| 172 |
+
step 1700000 | 9062s | train_loss=0.0323 | fid=239.2189
|
| 173 |
+
step 1710000 | 9095s | train_loss=0.0326
|
| 174 |
+
step 1720000 | 9129s | train_loss=0.0308
|
| 175 |
+
step 1730000 | 9168s | train_loss=0.0366
|
| 176 |
+
step 1740000 | 9207s | train_loss=0.0328
|
| 177 |
+
step 1750000 | 9246s | train_loss=0.0342
|
| 178 |
+
step 1760000 | 9286s | train_loss=0.0351
|
| 179 |
+
step 1770000 | 9325s | train_loss=0.0317
|
| 180 |
+
step 1780000 | 9364s | train_loss=0.0349
|
| 181 |
+
step 1790000 | 9403s | train_loss=0.0326
|
| 182 |
+
step 1800000 | 9478s | train_loss=0.0328 | fid=241.4806
|
| 183 |
+
step 1810000 | 9509s | train_loss=0.0314
|
| 184 |
+
step 1820000 | 9549s | train_loss=0.0331
|
| 185 |
+
step 1830000 | 9588s | train_loss=0.0342
|
| 186 |
+
step 1840000 | 9627s | train_loss=0.0311
|
| 187 |
+
step 1850000 | 9666s | train_loss=0.0332
|
| 188 |
+
step 1860000 | 9706s | train_loss=0.0335
|
| 189 |
+
step 1870000 | 9745s | train_loss=0.0316
|
| 190 |
+
step 1880000 | 9784s | train_loss=0.0317
|
| 191 |
+
step 1890000 | 9823s | train_loss=0.0339
|
| 192 |
+
step 1900000 | 9897s | train_loss=0.0323 | fid=244.6377
|
| 193 |
+
step 1910000 | 9934s | train_loss=0.0348
|
| 194 |
+
step 1920000 | 9973s | train_loss=0.0298
|
| 195 |
+
step 1930000 | 10012s | train_loss=0.0295
|
| 196 |
+
step 1940000 | 10051s | train_loss=0.0326
|
| 197 |
+
step 1950000 | 10090s | train_loss=0.0311
|
| 198 |
+
step 1960000 | 10130s | train_loss=0.0282
|
| 199 |
+
step 1970000 | 10168s | train_loss=0.0321
|
| 200 |
+
step 1980000 | 10208s | train_loss=0.0347
|
| 201 |
+
step 1990000 | 10247s | train_loss=0.0307
|
| 202 |
+
step 2000000 | 10309s | train_loss=0.0293 | fid=241.8694
|
| 203 |
+
step 2010000 | 10348s | train_loss=0.0341
|
| 204 |
+
step 2020000 | 10387s | train_loss=0.0368
|
| 205 |
+
step 2030000 | 10426s | train_loss=0.0306
|
| 206 |
+
step 2040000 | 10465s | train_loss=0.0350
|
| 207 |
+
step 2050000 | 10505s | train_loss=0.0359
|
| 208 |
+
step 2060000 | 10545s | train_loss=0.0360
|
| 209 |
+
step 2070000 | 10584s | train_loss=0.0315
|
| 210 |
+
step 2080000 | 10624s | train_loss=0.0313
|
| 211 |
+
step 2090000 | 10663s | train_loss=0.0316
|
| 212 |
+
step 2100000 | 10723s | train_loss=0.0345 | fid=236.9745
|
| 213 |
+
step 2110000 | 10762s | train_loss=0.0299
|
| 214 |
+
step 2120000 | 10801s | train_loss=0.0317
|
| 215 |
+
step 2130000 | 10840s | train_loss=0.0333
|
| 216 |
+
step 2140000 | 10880s | train_loss=0.0360
|
| 217 |
+
step 2150000 | 10920s | train_loss=0.0277
|
| 218 |
+
step 2160000 | 10959s | train_loss=0.0317
|
| 219 |
+
step 2170000 | 10998s | train_loss=0.0338
|
| 220 |
+
step 2180000 | 11037s | train_loss=0.0294
|
| 221 |
+
step 2190000 | 11076s | train_loss=0.0310
|
| 222 |
+
step 2200000 | 11139s | train_loss=0.0276 | fid=236.1073
|
| 223 |
+
step 2210000 | 11178s | train_loss=0.0312
|
| 224 |
+
step 2220000 | 11217s | train_loss=0.0367
|
| 225 |
+
step 2230000 | 11256s | train_loss=0.0294
|
| 226 |
+
step 2240000 | 11296s | train_loss=0.0297
|
| 227 |
+
step 2250000 | 11336s | train_loss=0.0306
|
| 228 |
+
step 2260000 | 11375s | train_loss=0.0345
|
| 229 |
+
step 2270000 | 11414s | train_loss=0.0375
|
| 230 |
+
step 2280000 | 11453s | train_loss=0.0355
|
| 231 |
+
step 2290000 | 11492s | train_loss=0.0305
|
| 232 |
+
step 2300000 | 11555s | train_loss=0.0290 | fid=234.0021
|
| 233 |
+
step 2310000 | 11594s | train_loss=0.0306
|
| 234 |
+
step 2320000 | 11634s | train_loss=0.0307
|
| 235 |
+
step 2330000 | 11673s | train_loss=0.0287
|
| 236 |
+
step 2340000 | 11712s | train_loss=0.0308
|
| 237 |
+
step 2350000 | 11752s | train_loss=0.0317
|
| 238 |
+
step 2360000 | 11791s | train_loss=0.0342
|
| 239 |
+
step 2370000 | 11830s | train_loss=0.0317
|
| 240 |
+
step 2380000 | 11869s | train_loss=0.0283
|
| 241 |
+
step 2390000 | 11900s | train_loss=0.0348
|
| 242 |
+
step 2400000 | 11969s | train_loss=0.0310 | fid=221.7723
|
| 243 |
+
step 2410000 | 12008s | train_loss=0.0274
|
| 244 |
+
step 2420000 | 12047s | train_loss=0.0299
|
| 245 |
+
step 2430000 | 12087s | train_loss=0.0305
|
| 246 |
+
step 2440000 | 12126s | train_loss=0.0278
|
| 247 |
+
step 2450000 | 12165s | train_loss=0.0328
|
| 248 |
+
step 2460000 | 12204s | train_loss=0.0297
|
| 249 |
+
step 2470000 | 12243s | train_loss=0.0302
|
| 250 |
+
step 2480000 | 12282s | train_loss=0.0274
|
| 251 |
+
step 2490000 | 12307s | train_loss=0.0282
|
| 252 |
+
step 2500000 | 12382s | train_loss=0.0276 | fid=226.8612
|
| 253 |
+
step 2510000 | 12422s | train_loss=0.0265
|
| 254 |
+
step 2520000 | 12461s | train_loss=0.0314
|
| 255 |
+
step 2530000 | 12501s | train_loss=0.0306
|
| 256 |
+
step 2540000 | 12540s | train_loss=0.0310
|
| 257 |
+
step 2550000 | 12579s | train_loss=0.0312
|
| 258 |
+
step 2560000 | 12619s | train_loss=0.0294
|
| 259 |
+
step 2570000 | 12658s | train_loss=0.0267
|
| 260 |
+
step 2580000 | 12696s | train_loss=0.0287
|
| 261 |
+
step 2590000 | 12726s | train_loss=0.0272
|
| 262 |
+
step 2600000 | 12806s | train_loss=0.0299 | fid=228.3404
|
| 263 |
+
step 2610000 | 12845s | train_loss=0.0318
|
| 264 |
+
step 2620000 | 12885s | train_loss=0.0282
|
| 265 |
+
step 2630000 | 12925s | train_loss=0.0293
|
| 266 |
+
step 2640000 | 12964s | train_loss=0.0291
|
| 267 |
+
step 2650000 | 13004s | train_loss=0.0295
|
| 268 |
+
step 2660000 | 13043s | train_loss=0.0268
|
| 269 |
+
step 2670000 | 13081s | train_loss=0.0283
|
| 270 |
+
step 2680000 | 13106s | train_loss=0.0297
|
| 271 |
+
step 2690000 | 13146s | train_loss=0.0299
|
| 272 |
+
step 2700000 | 13221s | train_loss=0.0285 | fid=223.4904
|
| 273 |
+
step 2710000 | 13260s | train_loss=0.0388
|
| 274 |
+
step 2720000 | 13299s | train_loss=0.0282
|
| 275 |
+
step 2730000 | 13339s | train_loss=0.0255
|
| 276 |
+
step 2740000 | 13378s | train_loss=0.0349
|
| 277 |
+
step 2750000 | 13417s | train_loss=0.0258
|
| 278 |
+
step 2760000 | 13456s | train_loss=0.0260
|
| 279 |
+
step 2770000 | 13493s | train_loss=0.0257
|
| 280 |
+
step 2780000 | 13522s | train_loss=0.0263
|
| 281 |
+
step 2790000 | 13561s | train_loss=0.0278
|
| 282 |
+
step 2800000 | 13636s | train_loss=0.0267 | fid=224.2581
|
| 283 |
+
step 2810000 | 13676s | train_loss=0.0281
|
| 284 |
+
step 2820000 | 13715s | train_loss=0.0310
|
| 285 |
+
step 2830000 | 13755s | train_loss=0.0303
|
| 286 |
+
step 2840000 | 13794s | train_loss=0.0268
|
| 287 |
+
step 2850000 | 13833s | train_loss=0.0293
|
| 288 |
+
step 2860000 | 13872s | train_loss=0.0256
|
| 289 |
+
step 2870000 | 13901s | train_loss=0.0289
|
| 290 |
+
step 2880000 | 13937s | train_loss=0.0300
|
| 291 |
+
step 2890000 | 13976s | train_loss=0.0301
|
| 292 |
+
step 2900000 | 14051s | train_loss=0.0254 | fid=218.6316
|
| 293 |
+
step 2910000 | 14090s | train_loss=0.0269
|
| 294 |
+
step 2920000 | 14130s | train_loss=0.0279
|
| 295 |
+
step 2930000 | 14169s | train_loss=0.0259
|
| 296 |
+
step 2940000 | 14209s | train_loss=0.0258
|
| 297 |
+
step 2950000 | 14248s | train_loss=0.0293
|
| 298 |
+
step 2960000 | 14286s | train_loss=0.0249
|
| 299 |
+
step 2970000 | 14326s | train_loss=0.0267
|
| 300 |
+
step 2980000 | 14365s | train_loss=0.0280
|
| 301 |
+
step 2990000 | 14404s | train_loss=0.0284
|
| 302 |
+
step 3000000 | 14479s | train_loss=0.0275 | fid=219.8106
|
| 303 |
+
step 3010000 | 14519s | train_loss=0.0288
|
| 304 |
+
step 3020000 | 14558s | train_loss=0.0260
|
| 305 |
+
step 3030000 | 14597s | train_loss=0.0353
|
| 306 |
+
step 3040000 | 14636s | train_loss=0.0268
|
| 307 |
+
step 3050000 | 14673s | train_loss=0.0285
|
| 308 |
+
step 3060000 | 14704s | train_loss=0.0268
|
| 309 |
+
step 3070000 | 14743s | train_loss=0.0273
|
| 310 |
+
step 3080000 | 14783s | train_loss=0.0282
|
| 311 |
+
step 3090000 | 14822s | train_loss=0.0327
|
| 312 |
+
step 3100000 | 14896s | train_loss=0.0331 | fid=227.9214
|
| 313 |
+
step 3110000 | 14935s | train_loss=0.0276
|
| 314 |
+
step 3120000 | 14974s | train_loss=0.0302
|
| 315 |
+
step 3130000 | 15014s | train_loss=0.0339
|
| 316 |
+
step 3140000 | 15053s | train_loss=0.0313
|
| 317 |
+
step 3150000 | 15092s | train_loss=0.0254
|
| 318 |
+
step 3160000 | 15131s | train_loss=0.0293
|
| 319 |
+
step 3170000 | 15171s | train_loss=0.0263
|
| 320 |
+
step 3180000 | 15210s | train_loss=0.0269
|
| 321 |
+
step 3190000 | 15249s | train_loss=0.0297
|
| 322 |
+
step 3200000 | 15323s | train_loss=0.0258 | fid=216.3038
|
| 323 |
+
step 3210000 | 15362s | train_loss=0.0306
|
| 324 |
+
step 3220000 | 15401s | train_loss=0.0250
|
| 325 |
+
step 3230000 | 15440s | train_loss=0.0259
|
| 326 |
+
step 3240000 | 15480s | train_loss=0.0265
|
| 327 |
+
step 3250000 | 15518s | train_loss=0.0316
|
| 328 |
+
step 3260000 | 15558s | train_loss=0.0252
|
| 329 |
+
step 3270000 | 15597s | train_loss=0.0351
|
| 330 |
+
step 3280000 | 15636s | train_loss=0.0345
|
| 331 |
+
step 3290000 | 15675s | train_loss=0.0274
|
| 332 |
+
step 3300000 | 15750s | train_loss=0.0273 | fid=216.3465
|
| 333 |
+
step 3310000 | 15789s | train_loss=0.0254
|
| 334 |
+
step 3320000 | 15828s | train_loss=0.0266
|
| 335 |
+
step 3330000 | 15868s | train_loss=0.0259
|
| 336 |
+
step 3340000 | 15907s | train_loss=0.0278
|
| 337 |
+
step 3350000 | 15947s | train_loss=0.0250
|
| 338 |
+
step 3360000 | 15986s | train_loss=0.0259
|
| 339 |
+
step 3370000 | 16026s | train_loss=0.0283
|
| 340 |
+
step 3380000 | 16065s | train_loss=0.0275
|
| 341 |
+
step 3390000 | 16104s | train_loss=0.0290
|
| 342 |
+
step 3400000 | 16179s | train_loss=0.0254 | fid=229.1554
|
| 343 |
+
step 3410000 | 16218s | train_loss=0.0273
|
| 344 |
+
step 3420000 | 16257s | train_loss=0.0269
|
| 345 |
+
step 3430000 | 16297s | train_loss=0.0273
|
| 346 |
+
step 3440000 | 16336s | train_loss=0.0292
|
| 347 |
+
step 3450000 | 16375s | train_loss=0.0244
|
| 348 |
+
step 3460000 | 16414s | train_loss=0.0244
|
| 349 |
+
step 3470000 | 16453s | train_loss=0.0247
|
| 350 |
+
step 3480000 | 16493s | train_loss=0.0245
|
| 351 |
+
step 3490000 | 16533s | train_loss=0.0261
|
| 352 |
+
step 3500000 | 16607s | train_loss=0.0254 | fid=220.9337
|
| 353 |
+
step 3510000 | 16646s | train_loss=0.0264
|
| 354 |
+
step 3520000 | 16685s | train_loss=0.0264
|
| 355 |
+
step 3530000 | 16725s | train_loss=0.0278
|
| 356 |
+
step 3540000 | 16764s | train_loss=0.0238
|
| 357 |
+
step 3550000 | 16803s | train_loss=0.0270
|
| 358 |
+
step 3560000 | 16842s | train_loss=0.0249
|
| 359 |
+
step 3570000 | 16882s | train_loss=0.0264
|
| 360 |
+
step 3580000 | 16921s | train_loss=0.0283
|
| 361 |
+
step 3590000 | 16961s | train_loss=0.0315
|
| 362 |
+
step 3600000 | 17036s | train_loss=0.0286 | fid=211.2366
|
| 363 |
+
step 3610000 | 17075s | train_loss=0.0277
|
| 364 |
+
step 3620000 | 17114s | train_loss=0.0271
|
| 365 |
+
step 3630000 | 17153s | train_loss=0.0285
|
| 366 |
+
step 3640000 | 17192s | train_loss=0.0282
|
| 367 |
+
step 3650000 | 17231s | train_loss=0.0316
|
| 368 |
+
step 3660000 | 17270s | train_loss=0.0263
|
| 369 |
+
step 3670000 | 17309s | train_loss=0.0261
|
| 370 |
+
step 3680000 | 17348s | train_loss=0.0270
|
| 371 |
+
step 3690000 | 17387s | train_loss=0.0268
|
| 372 |
+
step 3700000 | 17462s | train_loss=0.0252 | fid=213.4131
|
| 373 |
+
step 3710000 | 17501s | train_loss=0.0254
|
| 374 |
+
step 3720000 | 17540s | train_loss=0.0325
|
| 375 |
+
step 3730000 | 17580s | train_loss=0.0268
|
| 376 |
+
step 3740000 | 17619s | train_loss=0.0290
|
| 377 |
+
step 3750000 | 17658s | train_loss=0.0252
|
| 378 |
+
step 3760000 | 17697s | train_loss=0.0260
|
| 379 |
+
step 3770000 | 17735s | train_loss=0.0280
|
| 380 |
+
step 3780000 | 17775s | train_loss=0.0250
|
| 381 |
+
step 3790000 | 17815s | train_loss=0.0262
|
| 382 |
+
step 3800000 | 17890s | train_loss=0.0253 | fid=210.2569
|
| 383 |
+
step 3810000 | 17929s | train_loss=0.0285
|
| 384 |
+
step 3820000 | 17969s | train_loss=0.0244
|
| 385 |
+
step 3830000 | 18008s | train_loss=0.0265
|
| 386 |
+
step 3840000 | 18048s | train_loss=0.0248
|
| 387 |
+
step 3850000 | 18087s | train_loss=0.0253
|
| 388 |
+
step 3860000 | 18126s | train_loss=0.0260
|
| 389 |
+
step 3870000 | 18166s | train_loss=0.0244
|
| 390 |
+
step 3880000 | 18205s | train_loss=0.0248
|
| 391 |
+
step 3890000 | 18244s | train_loss=0.0229
|
| 392 |
+
step 3900000 | 18318s | train_loss=0.0250 | fid=216.1908
|
| 393 |
+
step 3910000 | 18357s | train_loss=0.0270
|
| 394 |
+
step 3920000 | 18396s | train_loss=0.0258
|
| 395 |
+
step 3930000 | 18436s | train_loss=0.0252
|
| 396 |
+
step 3940000 | 18474s | train_loss=0.0268
|
| 397 |
+
step 3950000 | 18513s | train_loss=0.0257
|
| 398 |
+
step 3960000 | 18553s | train_loss=0.0250
|
| 399 |
+
step 3970000 | 18592s | train_loss=0.0291
|
| 400 |
+
step 3980000 | 18632s | train_loss=0.0230
|
| 401 |
+
step 3990000 | 18671s | train_loss=0.0318
|
| 402 |
+
step 4000000 | 18746s | train_loss=0.0251 | fid=210.7833
|
| 403 |
+
step 4010000 | 18785s | train_loss=0.0253
|
| 404 |
+
step 4020000 | 18824s | train_loss=0.0261
|
| 405 |
+
step 4030000 | 18862s | train_loss=0.0281
|
| 406 |
+
step 4040000 | 18902s | train_loss=0.0263
|
| 407 |
+
step 4050000 | 18941s | train_loss=0.0247
|
| 408 |
+
step 4060000 | 18981s | train_loss=0.0240
|
| 409 |
+
step 4070000 | 19020s | train_loss=0.0231
|
| 410 |
+
step 4080000 | 19060s | train_loss=0.0285
|
| 411 |
+
step 4090000 | 19099s | train_loss=0.0265
|
| 412 |
+
step 4100000 | 19174s | train_loss=0.0257 | fid=208.8964
|
| 413 |
+
step 4110000 | 19213s | train_loss=0.0244
|
| 414 |
+
step 4120000 | 19252s | train_loss=0.0286
|
| 415 |
+
step 4130000 | 19291s | train_loss=0.0264
|
| 416 |
+
step 4140000 | 19330s | train_loss=0.0273
|
| 417 |
+
step 4150000 | 19370s | train_loss=0.0235
|
| 418 |
+
step 4160000 | 19409s | train_loss=0.0252
|
| 419 |
+
step 4170000 | 19448s | train_loss=0.0245
|
| 420 |
+
step 4180000 | 19487s | train_loss=0.0298
|
| 421 |
+
step 4190000 | 19526s | train_loss=0.0260
|
| 422 |
+
step 4200000 | 19601s | train_loss=0.0242 | fid=209.7965
|
| 423 |
+
step 4210000 | 19640s | train_loss=0.0266
|
| 424 |
+
step 4220000 | 19679s | train_loss=0.0283
|
| 425 |
+
step 4230000 | 19718s | train_loss=0.0231
|
| 426 |
+
step 4240000 | 19757s | train_loss=0.0235
|
| 427 |
+
step 4250000 | 19797s | train_loss=0.0275
|
| 428 |
+
step 4260000 | 19836s | train_loss=0.0227
|
| 429 |
+
step 4270000 | 19875s | train_loss=0.0239
|
| 430 |
+
step 4280000 | 19915s | train_loss=0.0234
|
| 431 |
+
step 4290000 | 19954s | train_loss=0.0247
|
| 432 |
+
step 4300000 | 20012s | train_loss=0.0243 | fid=207.7287
|
| 433 |
+
step 4310000 | 20051s | train_loss=0.0240
|
| 434 |
+
step 4320000 | 20091s | train_loss=0.0233
|
| 435 |
+
step 4330000 | 20131s | train_loss=0.0249
|
| 436 |
+
step 4340000 | 20171s | train_loss=0.0276
|
| 437 |
+
step 4350000 | 20211s | train_loss=0.0264
|
| 438 |
+
step 4360000 | 20250s | train_loss=0.0250
|
| 439 |
+
step 4370000 | 20290s | train_loss=0.0268
|
| 440 |
+
step 4380000 | 20329s | train_loss=0.0243
|
| 441 |
+
step 4390000 | 20369s | train_loss=0.0262
|
| 442 |
+
step 4400000 | 20431s | train_loss=0.0235 | fid=203.7425
|
| 443 |
+
step 4410000 | 20471s | train_loss=0.0255
|
| 444 |
+
step 4420000 | 20510s | train_loss=0.0235
|
| 445 |
+
step 4430000 | 20549s | train_loss=0.0281
|
| 446 |
+
step 4440000 | 20589s | train_loss=0.0237
|
| 447 |
+
step 4450000 | 20628s | train_loss=0.0245
|
| 448 |
+
step 4460000 | 20668s | train_loss=0.0264
|
| 449 |
+
step 4470000 | 20707s | train_loss=0.0223
|
| 450 |
+
step 4480000 | 20746s | train_loss=0.0229
|
| 451 |
+
step 4490000 | 20785s | train_loss=0.0250
|
| 452 |
+
step 4500000 | 20848s | train_loss=0.0236 | fid=207.5077
|
| 453 |
+
step 4510000 | 20887s | train_loss=0.0239
|
| 454 |
+
step 4520000 | 20927s | train_loss=0.0304
|
| 455 |
+
step 4530000 | 20966s | train_loss=0.0234
|
| 456 |
+
step 4540000 | 21006s | train_loss=0.0246
|
| 457 |
+
step 4550000 | 21045s | train_loss=0.0293
|
| 458 |
+
step 4560000 | 21084s | train_loss=0.0240
|
| 459 |
+
step 4570000 | 21124s | train_loss=0.0226
|
| 460 |
+
step 4580000 | 21164s | train_loss=0.0261
|
| 461 |
+
step 4590000 | 21202s | train_loss=0.0235
|
| 462 |
+
step 4600000 | 21277s | train_loss=0.0304 | fid=202.3799
|
| 463 |
+
step 4610000 | 21316s | train_loss=0.0225
|
| 464 |
+
step 4620000 | 21356s | train_loss=0.0266
|
| 465 |
+
step 4630000 | 21395s | train_loss=0.0276
|
| 466 |
+
step 4640000 | 21435s | train_loss=0.0250
|
| 467 |
+
step 4650000 | 21474s | train_loss=0.0225
|
| 468 |
+
step 4660000 | 21513s | train_loss=0.0259
|
| 469 |
+
step 4670000 | 21551s | train_loss=0.0239
|
| 470 |
+
step 4680000 | 21579s | train_loss=0.0230
|
| 471 |
+
step 4690000 | 21602s | train_loss=0.0248
|
| 472 |
+
step 4700000 | 21644s | train_loss=0.0251 | fid=197.1110
|
| 473 |
+
step 4710000 | 21666s | train_loss=0.0215
|
| 474 |
+
step 4720000 | 21689s | train_loss=0.0229
|
| 475 |
+
step 4730000 | 21712s | train_loss=0.0231
|
| 476 |
+
step 4740000 | 21735s | train_loss=0.0246
|
| 477 |
+
step 4750000 | 21758s | train_loss=0.0226
|
| 478 |
+
step 4760000 | 21781s | train_loss=0.0228
|
| 479 |
+
step 4770000 | 21804s | train_loss=0.0288
|
| 480 |
+
step 4780000 | 21827s | train_loss=0.0247
|
| 481 |
+
step 4790000 | 21850s | train_loss=0.0257
|
| 482 |
+
step 4800000 | 21891s | train_loss=0.0233 | fid=210.1880
|
| 483 |
+
step 4810000 | 21914s | train_loss=0.0227
|
| 484 |
+
step 4820000 | 21937s | train_loss=0.0230
|
| 485 |
+
step 4830000 | 21960s | train_loss=0.0268
|
| 486 |
+
step 4840000 | 21983s | train_loss=0.0228
|
| 487 |
+
step 4850000 | 22006s | train_loss=0.0215
|
| 488 |
+
step 4860000 | 22029s | train_loss=0.0258
|
| 489 |
+
step 4870000 | 22052s | train_loss=0.0214
|
| 490 |
+
step 4880000 | 22075s | train_loss=0.0229
|
| 491 |
+
step 4890000 | 22098s | train_loss=0.0226
|
| 492 |
+
step 4900000 | 22139s | train_loss=0.0279 | fid=203.6338
|
| 493 |
+
step 4910000 | 22162s | train_loss=0.0226
|
| 494 |
+
step 4920000 | 22183s | train_loss=0.0240
|
| 495 |
+
step 4930000 | 22205s | train_loss=0.0231
|
| 496 |
+
step 4940000 | 22228s | train_loss=0.0226
|
| 497 |
+
step 4950000 | 22249s | train_loss=0.0245
|
| 498 |
+
step 4960000 | 22272s | train_loss=0.0238
|
| 499 |
+
step 4970000 | 22292s | train_loss=0.0232
|
| 500 |
+
step 4980000 | 22315s | train_loss=0.0229
|
| 501 |
+
step 4990000 | 22338s | train_loss=0.0227
|
| 502 |
+
step 5000000 | 22379s | train_loss=0.0229 | fid=198.8955
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d240/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d240/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d240",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 44,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d240/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d240/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 47s | train_loss=1.0045 | fid=298.2617
|
| 3 |
+
step 10000 | 100s | train_loss=0.7634
|
| 4 |
+
step 20000 | 156s | train_loss=0.6588
|
| 5 |
+
step 30000 | 212s | train_loss=0.6235
|
| 6 |
+
step 40000 | 268s | train_loss=0.5755
|
| 7 |
+
step 50000 | 324s | train_loss=0.5472
|
| 8 |
+
step 60000 | 380s | train_loss=0.5121
|
| 9 |
+
step 70000 | 436s | train_loss=0.4760
|
| 10 |
+
step 80000 | 492s | train_loss=0.4527
|
| 11 |
+
step 90000 | 547s | train_loss=0.4290
|
| 12 |
+
step 100000 | 651s | train_loss=0.4023 | fid=142.1030
|
| 13 |
+
step 110000 | 703s | train_loss=0.3810
|
| 14 |
+
step 120000 | 759s | train_loss=0.3585
|
| 15 |
+
step 130000 | 815s | train_loss=0.3330
|
| 16 |
+
step 140000 | 871s | train_loss=0.3117
|
| 17 |
+
step 150000 | 927s | train_loss=0.2947
|
| 18 |
+
step 160000 | 983s | train_loss=0.2787
|
| 19 |
+
step 170000 | 1039s | train_loss=0.2657
|
| 20 |
+
step 180000 | 1094s | train_loss=0.2497
|
| 21 |
+
step 190000 | 1150s | train_loss=0.2354
|
| 22 |
+
step 200000 | 1256s | train_loss=0.2285 | fid=149.7584
|
| 23 |
+
step 210000 | 1313s | train_loss=0.2159
|
| 24 |
+
step 220000 | 1365s | train_loss=0.2074
|
| 25 |
+
step 230000 | 1421s | train_loss=0.1973
|
| 26 |
+
step 240000 | 1477s | train_loss=0.1873
|
| 27 |
+
step 250000 | 1533s | train_loss=0.1840
|
| 28 |
+
step 260000 | 1589s | train_loss=0.1790
|
| 29 |
+
step 270000 | 1645s | train_loss=0.1689
|
| 30 |
+
step 280000 | 1701s | train_loss=0.1702
|
| 31 |
+
step 290000 | 1757s | train_loss=0.1671
|
| 32 |
+
step 300000 | 1863s | train_loss=0.1598 | fid=155.7782
|
| 33 |
+
step 310000 | 1919s | train_loss=0.1553
|
| 34 |
+
step 320000 | 1971s | train_loss=0.1482
|
| 35 |
+
step 330000 | 2027s | train_loss=0.1473
|
| 36 |
+
step 340000 | 2083s | train_loss=0.1446
|
| 37 |
+
step 350000 | 2139s | train_loss=0.1394
|
| 38 |
+
step 360000 | 2195s | train_loss=0.1359
|
| 39 |
+
step 370000 | 2251s | train_loss=0.1351
|
| 40 |
+
step 380000 | 2307s | train_loss=0.1318
|
| 41 |
+
step 390000 | 2363s | train_loss=0.1287
|
| 42 |
+
step 400000 | 2466s | train_loss=0.1294 | fid=150.4909
|
| 43 |
+
step 410000 | 2522s | train_loss=0.1261
|
| 44 |
+
step 420000 | 2578s | train_loss=0.1205
|
| 45 |
+
step 430000 | 2631s | train_loss=0.1174
|
| 46 |
+
step 440000 | 2686s | train_loss=0.1184
|
| 47 |
+
step 450000 | 2742s | train_loss=0.1158
|
| 48 |
+
step 460000 | 2798s | train_loss=0.1144
|
| 49 |
+
step 470000 | 2854s | train_loss=0.1128
|
| 50 |
+
step 480000 | 2910s | train_loss=0.1108
|
| 51 |
+
step 490000 | 2966s | train_loss=0.1103
|
| 52 |
+
step 500000 | 3070s | train_loss=0.1101 | fid=146.3849
|
| 53 |
+
step 510000 | 3126s | train_loss=0.1050
|
| 54 |
+
step 520000 | 3181s | train_loss=0.1048
|
| 55 |
+
step 530000 | 3236s | train_loss=0.1051
|
| 56 |
+
step 540000 | 3291s | train_loss=0.1007
|
| 57 |
+
step 550000 | 3347s | train_loss=0.1010
|
| 58 |
+
step 560000 | 3403s | train_loss=0.0972
|
| 59 |
+
step 570000 | 3459s | train_loss=0.0938
|
| 60 |
+
step 580000 | 3515s | train_loss=0.0940
|
| 61 |
+
step 590000 | 3571s | train_loss=0.0919
|
| 62 |
+
step 600000 | 3674s | train_loss=0.0907 | fid=151.8681
|
| 63 |
+
step 610000 | 3730s | train_loss=0.0912
|
| 64 |
+
step 620000 | 3786s | train_loss=0.0879
|
| 65 |
+
step 630000 | 3842s | train_loss=0.0874
|
| 66 |
+
step 640000 | 3895s | train_loss=0.0867
|
| 67 |
+
step 650000 | 3951s | train_loss=0.0891
|
| 68 |
+
step 660000 | 4007s | train_loss=0.0848
|
| 69 |
+
step 670000 | 4063s | train_loss=0.0826
|
| 70 |
+
step 680000 | 4119s | train_loss=0.0788
|
| 71 |
+
step 690000 | 4175s | train_loss=0.0824
|
| 72 |
+
step 700000 | 4279s | train_loss=0.0778 | fid=172.5369
|
| 73 |
+
step 710000 | 4335s | train_loss=0.0813
|
| 74 |
+
step 720000 | 4391s | train_loss=0.0783
|
| 75 |
+
step 730000 | 4446s | train_loss=0.0782
|
| 76 |
+
step 740000 | 4503s | train_loss=0.0738
|
| 77 |
+
step 750000 | 4555s | train_loss=0.0738
|
| 78 |
+
step 760000 | 4611s | train_loss=0.0746
|
| 79 |
+
step 770000 | 4667s | train_loss=0.0737
|
| 80 |
+
step 780000 | 4723s | train_loss=0.0764
|
| 81 |
+
step 790000 | 4779s | train_loss=0.0705
|
| 82 |
+
step 800000 | 4883s | train_loss=0.0687 | fid=190.7247
|
| 83 |
+
step 810000 | 4939s | train_loss=0.0714
|
| 84 |
+
step 820000 | 4995s | train_loss=0.0703
|
| 85 |
+
step 830000 | 5051s | train_loss=0.0686
|
| 86 |
+
step 840000 | 5106s | train_loss=0.0688
|
| 87 |
+
step 850000 | 5160s | train_loss=0.0676
|
| 88 |
+
step 860000 | 5216s | train_loss=0.0674
|
| 89 |
+
step 870000 | 5272s | train_loss=0.0652
|
| 90 |
+
step 880000 | 5328s | train_loss=0.0666
|
| 91 |
+
step 890000 | 5384s | train_loss=0.0656
|
| 92 |
+
step 900000 | 5488s | train_loss=0.0648 | fid=208.0946
|
| 93 |
+
step 910000 | 5544s | train_loss=0.0665
|
| 94 |
+
step 920000 | 5586s | train_loss=0.0613
|
| 95 |
+
step 930000 | 5622s | train_loss=0.0638
|
| 96 |
+
step 940000 | 5659s | train_loss=0.0630
|
| 97 |
+
step 950000 | 5696s | train_loss=0.0612
|
| 98 |
+
step 960000 | 5728s | train_loss=0.0661
|
| 99 |
+
step 970000 | 5765s | train_loss=0.0630
|
| 100 |
+
step 980000 | 5802s | train_loss=0.0633
|
| 101 |
+
step 990000 | 5839s | train_loss=0.0599
|
| 102 |
+
step 1000000 | 5912s | train_loss=0.0603 | fid=207.9026
|
| 103 |
+
step 1010000 | 5949s | train_loss=0.0601
|
| 104 |
+
step 1020000 | 5986s | train_loss=0.0585
|
| 105 |
+
step 1030000 | 6023s | train_loss=0.0752
|
| 106 |
+
step 1040000 | 6060s | train_loss=0.0583
|
| 107 |
+
step 1050000 | 6097s | train_loss=0.0573
|
| 108 |
+
step 1060000 | 6132s | train_loss=0.0589
|
| 109 |
+
step 1070000 | 6167s | train_loss=0.0575
|
| 110 |
+
step 1080000 | 6203s | train_loss=0.0579
|
| 111 |
+
step 1090000 | 6240s | train_loss=0.0582
|
| 112 |
+
step 1100000 | 6310s | train_loss=0.0546 | fid=212.5103
|
| 113 |
+
step 1110000 | 6347s | train_loss=0.0590
|
| 114 |
+
step 1120000 | 6384s | train_loss=0.0528
|
| 115 |
+
step 1130000 | 6421s | train_loss=0.0603
|
| 116 |
+
step 1140000 | 6457s | train_loss=0.0534
|
| 117 |
+
step 1150000 | 6494s | train_loss=0.0533
|
| 118 |
+
step 1160000 | 6531s | train_loss=0.0556
|
| 119 |
+
step 1170000 | 6563s | train_loss=0.0541
|
| 120 |
+
step 1180000 | 6600s | train_loss=0.0532
|
| 121 |
+
step 1190000 | 6637s | train_loss=0.0520
|
| 122 |
+
step 1200000 | 6707s | train_loss=0.0519 | fid=214.7850
|
| 123 |
+
step 1210000 | 6744s | train_loss=0.0519
|
| 124 |
+
step 1220000 | 6781s | train_loss=0.0564
|
| 125 |
+
step 1230000 | 6818s | train_loss=0.0537
|
| 126 |
+
step 1240000 | 6855s | train_loss=0.0501
|
| 127 |
+
step 1250000 | 6892s | train_loss=0.0580
|
| 128 |
+
step 1260000 | 6929s | train_loss=0.0492
|
| 129 |
+
step 1270000 | 6965s | train_loss=0.0527
|
| 130 |
+
step 1280000 | 6998s | train_loss=0.0518
|
| 131 |
+
step 1290000 | 7035s | train_loss=0.0496
|
| 132 |
+
step 1300000 | 7105s | train_loss=0.0536 | fid=207.2809
|
| 133 |
+
step 1310000 | 7142s | train_loss=0.0480
|
| 134 |
+
step 1320000 | 7179s | train_loss=0.0517
|
| 135 |
+
step 1330000 | 7216s | train_loss=0.0496
|
| 136 |
+
step 1340000 | 7253s | train_loss=0.0495
|
| 137 |
+
step 1350000 | 7290s | train_loss=0.0510
|
| 138 |
+
step 1360000 | 7327s | train_loss=0.0515
|
| 139 |
+
step 1370000 | 7364s | train_loss=0.0456
|
| 140 |
+
step 1380000 | 7394s | train_loss=0.0470
|
| 141 |
+
step 1390000 | 7431s | train_loss=0.0505
|
| 142 |
+
step 1400000 | 7502s | train_loss=0.0513 | fid=201.1736
|
| 143 |
+
step 1410000 | 7539s | train_loss=0.0476
|
| 144 |
+
step 1420000 | 7576s | train_loss=0.0544
|
| 145 |
+
step 1430000 | 7613s | train_loss=0.0463
|
| 146 |
+
step 1440000 | 7650s | train_loss=0.0460
|
| 147 |
+
step 1450000 | 7687s | train_loss=0.0514
|
| 148 |
+
step 1460000 | 7724s | train_loss=0.0505
|
| 149 |
+
step 1470000 | 7761s | train_loss=0.0458
|
| 150 |
+
step 1480000 | 7798s | train_loss=0.0475
|
| 151 |
+
step 1490000 | 7829s | train_loss=0.0464
|
| 152 |
+
step 1500000 | 7900s | train_loss=0.0444 | fid=202.8455
|
| 153 |
+
step 1510000 | 7937s | train_loss=0.0454
|
| 154 |
+
step 1520000 | 7974s | train_loss=0.0522
|
| 155 |
+
step 1530000 | 8011s | train_loss=0.0477
|
| 156 |
+
step 1540000 | 8047s | train_loss=0.0470
|
| 157 |
+
step 1550000 | 8084s | train_loss=0.0480
|
| 158 |
+
step 1560000 | 8121s | train_loss=0.0483
|
| 159 |
+
step 1570000 | 8158s | train_loss=0.0422
|
| 160 |
+
step 1580000 | 8195s | train_loss=0.0457
|
| 161 |
+
step 1590000 | 8228s | train_loss=0.0452
|
| 162 |
+
step 1600000 | 8299s | train_loss=0.0461 | fid=189.5870
|
| 163 |
+
step 1610000 | 8336s | train_loss=0.0476
|
| 164 |
+
step 1620000 | 8373s | train_loss=0.0418
|
| 165 |
+
step 1630000 | 8410s | train_loss=0.0419
|
| 166 |
+
step 1640000 | 8447s | train_loss=0.0464
|
| 167 |
+
step 1650000 | 8484s | train_loss=0.0455
|
| 168 |
+
step 1660000 | 8521s | train_loss=0.0515
|
| 169 |
+
step 1670000 | 8558s | train_loss=0.0445
|
| 170 |
+
step 1680000 | 8594s | train_loss=0.0435
|
| 171 |
+
step 1690000 | 8631s | train_loss=0.0509
|
| 172 |
+
step 1700000 | 8698s | train_loss=0.0408 | fid=178.1438
|
| 173 |
+
step 1710000 | 8735s | train_loss=0.0414
|
| 174 |
+
step 1720000 | 8772s | train_loss=0.0397
|
| 175 |
+
step 1730000 | 8809s | train_loss=0.0443
|
| 176 |
+
step 1740000 | 8846s | train_loss=0.0427
|
| 177 |
+
step 1750000 | 8883s | train_loss=0.0438
|
| 178 |
+
step 1760000 | 8920s | train_loss=0.0438
|
| 179 |
+
step 1770000 | 8957s | train_loss=0.0408
|
| 180 |
+
step 1780000 | 8994s | train_loss=0.0426
|
| 181 |
+
step 1790000 | 9031s | train_loss=0.0403
|
| 182 |
+
step 1800000 | 9099s | train_loss=0.0409 | fid=185.6562
|
| 183 |
+
step 1810000 | 9136s | train_loss=0.0399
|
| 184 |
+
step 1820000 | 9173s | train_loss=0.0410
|
| 185 |
+
step 1830000 | 9210s | train_loss=0.0451
|
| 186 |
+
step 1840000 | 9247s | train_loss=0.0419
|
| 187 |
+
step 1850000 | 9284s | train_loss=0.0407
|
| 188 |
+
step 1860000 | 9321s | train_loss=0.0413
|
| 189 |
+
step 1870000 | 9357s | train_loss=0.0390
|
| 190 |
+
step 1880000 | 9394s | train_loss=0.0391
|
| 191 |
+
step 1890000 | 9431s | train_loss=0.0432
|
| 192 |
+
step 1900000 | 9493s | train_loss=0.0408 | fid=181.5280
|
| 193 |
+
step 1910000 | 9530s | train_loss=0.0466
|
| 194 |
+
step 1920000 | 9567s | train_loss=0.0376
|
| 195 |
+
step 1930000 | 9604s | train_loss=0.0373
|
| 196 |
+
step 1940000 | 9641s | train_loss=0.0411
|
| 197 |
+
step 1950000 | 9678s | train_loss=0.0387
|
| 198 |
+
step 1960000 | 9715s | train_loss=0.0367
|
| 199 |
+
step 1970000 | 9752s | train_loss=0.0382
|
| 200 |
+
step 1980000 | 9789s | train_loss=0.0449
|
| 201 |
+
step 1990000 | 9826s | train_loss=0.0358
|
| 202 |
+
step 2000000 | 9900s | train_loss=0.0385 | fid=169.4751
|
| 203 |
+
step 2010000 | 9937s | train_loss=0.0414
|
| 204 |
+
step 2020000 | 9974s | train_loss=0.0472
|
| 205 |
+
step 2030000 | 10011s | train_loss=0.0376
|
| 206 |
+
step 2040000 | 10048s | train_loss=0.0454
|
| 207 |
+
step 2050000 | 10085s | train_loss=0.0427
|
| 208 |
+
step 2060000 | 10122s | train_loss=0.0404
|
| 209 |
+
step 2070000 | 10159s | train_loss=0.0389
|
| 210 |
+
step 2080000 | 10196s | train_loss=0.0377
|
| 211 |
+
step 2090000 | 10233s | train_loss=0.0376
|
| 212 |
+
step 2100000 | 10303s | train_loss=0.0410 | fid=167.1597
|
| 213 |
+
step 2110000 | 10337s | train_loss=0.0361
|
| 214 |
+
step 2120000 | 10373s | train_loss=0.0398
|
| 215 |
+
step 2130000 | 10410s | train_loss=0.0410
|
| 216 |
+
step 2140000 | 10447s | train_loss=0.0429
|
| 217 |
+
step 2150000 | 10484s | train_loss=0.0346
|
| 218 |
+
step 2160000 | 10521s | train_loss=0.0385
|
| 219 |
+
step 2170000 | 10558s | train_loss=0.0353
|
| 220 |
+
step 2180000 | 10595s | train_loss=0.0369
|
| 221 |
+
step 2190000 | 10632s | train_loss=0.0364
|
| 222 |
+
step 2200000 | 10704s | train_loss=0.0341 | fid=167.5430
|
| 223 |
+
step 2210000 | 10736s | train_loss=0.0395
|
| 224 |
+
step 2220000 | 10773s | train_loss=0.0401
|
| 225 |
+
step 2230000 | 10810s | train_loss=0.0354
|
| 226 |
+
step 2240000 | 10846s | train_loss=0.0360
|
| 227 |
+
step 2250000 | 10884s | train_loss=0.0392
|
| 228 |
+
step 2260000 | 10921s | train_loss=0.0375
|
| 229 |
+
step 2270000 | 10958s | train_loss=0.0406
|
| 230 |
+
step 2280000 | 10994s | train_loss=0.0405
|
| 231 |
+
step 2290000 | 11031s | train_loss=0.0367
|
| 232 |
+
step 2300000 | 11104s | train_loss=0.0360 | fid=162.6257
|
| 233 |
+
step 2310000 | 11137s | train_loss=0.0368
|
| 234 |
+
step 2320000 | 11173s | train_loss=0.0372
|
| 235 |
+
step 2330000 | 11210s | train_loss=0.0353
|
| 236 |
+
step 2340000 | 11247s | train_loss=0.0374
|
| 237 |
+
step 2350000 | 11284s | train_loss=0.0385
|
| 238 |
+
step 2360000 | 11321s | train_loss=0.0397
|
| 239 |
+
step 2370000 | 11358s | train_loss=0.0386
|
| 240 |
+
step 2380000 | 11395s | train_loss=0.0342
|
| 241 |
+
step 2390000 | 11431s | train_loss=0.0377
|
| 242 |
+
step 2400000 | 11507s | train_loss=0.0361 | fid=169.4406
|
| 243 |
+
step 2410000 | 11544s | train_loss=0.0332
|
| 244 |
+
step 2420000 | 11572s | train_loss=0.0361
|
| 245 |
+
step 2430000 | 11609s | train_loss=0.0357
|
| 246 |
+
step 2440000 | 11646s | train_loss=0.0334
|
| 247 |
+
step 2450000 | 11683s | train_loss=0.0391
|
| 248 |
+
step 2460000 | 11720s | train_loss=0.0359
|
| 249 |
+
step 2470000 | 11757s | train_loss=0.0369
|
| 250 |
+
step 2480000 | 11794s | train_loss=0.0343
|
| 251 |
+
step 2490000 | 11831s | train_loss=0.0350
|
| 252 |
+
step 2500000 | 11906s | train_loss=0.0348 | fid=153.5021
|
| 253 |
+
step 2510000 | 11942s | train_loss=0.0322
|
| 254 |
+
step 2520000 | 11972s | train_loss=0.0362
|
| 255 |
+
step 2530000 | 12009s | train_loss=0.0349
|
| 256 |
+
step 2540000 | 12046s | train_loss=0.0366
|
| 257 |
+
step 2550000 | 12083s | train_loss=0.0363
|
| 258 |
+
step 2560000 | 12120s | train_loss=0.0339
|
| 259 |
+
step 2570000 | 12157s | train_loss=0.0330
|
| 260 |
+
step 2580000 | 12194s | train_loss=0.0339
|
| 261 |
+
step 2590000 | 12231s | train_loss=0.0346
|
| 262 |
+
step 2600000 | 12306s | train_loss=0.0341 | fid=154.5368
|
| 263 |
+
step 2610000 | 12342s | train_loss=0.0343
|
| 264 |
+
step 2620000 | 12377s | train_loss=0.0326
|
| 265 |
+
step 2630000 | 12411s | train_loss=0.0346
|
| 266 |
+
step 2640000 | 12448s | train_loss=0.0338
|
| 267 |
+
step 2650000 | 12485s | train_loss=0.0357
|
| 268 |
+
step 2660000 | 12522s | train_loss=0.0335
|
| 269 |
+
step 2670000 | 12559s | train_loss=0.0340
|
| 270 |
+
step 2680000 | 12596s | train_loss=0.0336
|
| 271 |
+
step 2690000 | 12633s | train_loss=0.0342
|
| 272 |
+
step 2700000 | 12705s | train_loss=0.0323 | fid=148.3220
|
| 273 |
+
step 2710000 | 12742s | train_loss=0.0413
|
| 274 |
+
step 2720000 | 12779s | train_loss=0.0334
|
| 275 |
+
step 2730000 | 12805s | train_loss=0.0299
|
| 276 |
+
step 2740000 | 12842s | train_loss=0.0411
|
| 277 |
+
step 2750000 | 12879s | train_loss=0.0312
|
| 278 |
+
step 2760000 | 12917s | train_loss=0.0306
|
| 279 |
+
step 2770000 | 12953s | train_loss=0.0309
|
| 280 |
+
step 2780000 | 12990s | train_loss=0.0316
|
| 281 |
+
step 2790000 | 13027s | train_loss=0.0328
|
| 282 |
+
step 2800000 | 13102s | train_loss=0.0326 | fid=148.2657
|
| 283 |
+
step 2810000 | 13139s | train_loss=0.0325
|
| 284 |
+
step 2820000 | 13176s | train_loss=0.0351
|
| 285 |
+
step 2830000 | 13212s | train_loss=0.0343
|
| 286 |
+
step 2840000 | 13245s | train_loss=0.0318
|
| 287 |
+
step 2850000 | 13282s | train_loss=0.0339
|
| 288 |
+
step 2860000 | 13319s | train_loss=0.0321
|
| 289 |
+
step 2870000 | 13356s | train_loss=0.0320
|
| 290 |
+
step 2880000 | 13392s | train_loss=0.0328
|
| 291 |
+
step 2890000 | 13429s | train_loss=0.0342
|
| 292 |
+
step 2900000 | 13504s | train_loss=0.0314 | fid=150.8561
|
| 293 |
+
step 2910000 | 13541s | train_loss=0.0327
|
| 294 |
+
step 2920000 | 13578s | train_loss=0.0335
|
| 295 |
+
step 2930000 | 13614s | train_loss=0.0314
|
| 296 |
+
step 2940000 | 13646s | train_loss=0.0312
|
| 297 |
+
step 2950000 | 13683s | train_loss=0.0346
|
| 298 |
+
step 2960000 | 13720s | train_loss=0.0301
|
| 299 |
+
step 2970000 | 13757s | train_loss=0.0309
|
| 300 |
+
step 2980000 | 13794s | train_loss=0.0333
|
| 301 |
+
step 2990000 | 13830s | train_loss=0.0322
|
| 302 |
+
step 3000000 | 13905s | train_loss=0.0319 | fid=144.0020
|
| 303 |
+
step 3010000 | 13942s | train_loss=0.0317
|
| 304 |
+
step 3020000 | 13979s | train_loss=0.0316
|
| 305 |
+
step 3030000 | 14015s | train_loss=0.0378
|
| 306 |
+
step 3040000 | 14049s | train_loss=0.0326
|
| 307 |
+
step 3050000 | 14085s | train_loss=0.0341
|
| 308 |
+
step 3060000 | 14122s | train_loss=0.0322
|
| 309 |
+
step 3070000 | 14159s | train_loss=0.0319
|
| 310 |
+
step 3080000 | 14196s | train_loss=0.0369
|
| 311 |
+
step 3090000 | 14233s | train_loss=0.0347
|
| 312 |
+
step 3100000 | 14284s | train_loss=0.0385 | fid=nan
|
| 313 |
+
step 3110000 | 14321s | train_loss=0.0351
|
| 314 |
+
step 3120000 | 14357s | train_loss=0.0335
|
| 315 |
+
step 3130000 | 14394s | train_loss=0.0371
|
| 316 |
+
step 3140000 | 14431s | train_loss=0.0368
|
| 317 |
+
step 3150000 | 14468s | train_loss=0.0315
|
| 318 |
+
step 3160000 | 14497s | train_loss=0.0331
|
| 319 |
+
step 3170000 | 14535s | train_loss=0.0328
|
| 320 |
+
step 3180000 | 14571s | train_loss=0.0328
|
| 321 |
+
step 3190000 | 14608s | train_loss=0.0341
|
| 322 |
+
step 3200000 | 14682s | train_loss=0.0301 | fid=152.7347
|
| 323 |
+
step 3210000 | 14719s | train_loss=0.0357
|
| 324 |
+
step 3220000 | 14756s | train_loss=0.0303
|
| 325 |
+
step 3230000 | 14793s | train_loss=0.0313
|
| 326 |
+
step 3240000 | 14830s | train_loss=0.0315
|
| 327 |
+
step 3250000 | 14867s | train_loss=0.0364
|
| 328 |
+
step 3260000 | 14898s | train_loss=0.0298
|
| 329 |
+
step 3270000 | 14935s | train_loss=0.0331
|
| 330 |
+
step 3280000 | 14972s | train_loss=0.0367
|
| 331 |
+
step 3290000 | 15009s | train_loss=0.0309
|
| 332 |
+
step 3300000 | 15060s | train_loss=0.0314 | fid=nan
|
| 333 |
+
step 3310000 | 15097s | train_loss=0.0294
|
| 334 |
+
step 3320000 | 15134s | train_loss=0.0307
|
| 335 |
+
step 3330000 | 15170s | train_loss=0.0293
|
| 336 |
+
step 3340000 | 15207s | train_loss=0.0351
|
| 337 |
+
step 3350000 | 15244s | train_loss=0.0303
|
| 338 |
+
step 3360000 | 15281s | train_loss=0.0300
|
| 339 |
+
step 3370000 | 15317s | train_loss=0.0307
|
| 340 |
+
step 3380000 | 15350s | train_loss=0.0317
|
| 341 |
+
step 3390000 | 15387s | train_loss=0.0334
|
| 342 |
+
step 3400000 | 15438s | train_loss=0.0309 | fid=nan
|
| 343 |
+
step 3410000 | 15475s | train_loss=0.0313
|
| 344 |
+
step 3420000 | 15512s | train_loss=0.0334
|
| 345 |
+
step 3430000 | 15549s | train_loss=0.0314
|
| 346 |
+
step 3440000 | 15586s | train_loss=0.0303
|
| 347 |
+
step 3450000 | 15622s | train_loss=0.0292
|
| 348 |
+
step 3460000 | 15659s | train_loss=0.0290
|
| 349 |
+
step 3470000 | 15696s | train_loss=0.0286
|
| 350 |
+
step 3480000 | 15733s | train_loss=0.0279
|
| 351 |
+
step 3490000 | 15766s | train_loss=0.0299
|
| 352 |
+
step 3500000 | 15817s | train_loss=0.0288 | fid=nan
|
| 353 |
+
step 3510000 | 15854s | train_loss=0.0304
|
| 354 |
+
step 3520000 | 15891s | train_loss=0.0294
|
| 355 |
+
step 3530000 | 15928s | train_loss=0.0310
|
| 356 |
+
step 3540000 | 15965s | train_loss=0.0288
|
| 357 |
+
step 3550000 | 16002s | train_loss=0.0313
|
| 358 |
+
step 3560000 | 16039s | train_loss=0.0303
|
| 359 |
+
step 3570000 | 16076s | train_loss=0.0302
|
| 360 |
+
step 3580000 | 16112s | train_loss=0.0292
|
| 361 |
+
step 3590000 | 16150s | train_loss=0.0304
|
| 362 |
+
step 3600000 | 16195s | train_loss=0.0318 | fid=nan
|
| 363 |
+
step 3610000 | 16232s | train_loss=0.0286
|
| 364 |
+
step 3620000 | 16269s | train_loss=0.0300
|
| 365 |
+
step 3630000 | 16306s | train_loss=0.0312
|
| 366 |
+
step 3640000 | 16343s | train_loss=0.0291
|
| 367 |
+
step 3650000 | 16380s | train_loss=0.0304
|
| 368 |
+
step 3660000 | 16417s | train_loss=0.0291
|
| 369 |
+
step 3670000 | 16453s | train_loss=0.0309
|
| 370 |
+
step 3680000 | 16490s | train_loss=0.0310
|
| 371 |
+
step 3690000 | 16527s | train_loss=0.0308
|
| 372 |
+
step 3700000 | 16579s | train_loss=0.0285 | fid=nan
|
| 373 |
+
step 3710000 | 16610s | train_loss=0.0288
|
| 374 |
+
step 3720000 | 16647s | train_loss=0.0338
|
| 375 |
+
step 3730000 | 16684s | train_loss=0.0293
|
| 376 |
+
step 3740000 | 16720s | train_loss=0.0309
|
| 377 |
+
step 3750000 | 16757s | train_loss=0.0304
|
| 378 |
+
step 3760000 | 16794s | train_loss=0.0289
|
| 379 |
+
step 3770000 | 16831s | train_loss=0.0288
|
| 380 |
+
step 3780000 | 16868s | train_loss=0.0282
|
| 381 |
+
step 3790000 | 16905s | train_loss=0.0293
|
| 382 |
+
step 3800000 | 16956s | train_loss=0.0284 | fid=nan
|
| 383 |
+
step 3810000 | 16993s | train_loss=0.0302
|
| 384 |
+
step 3820000 | 17029s | train_loss=0.0287
|
| 385 |
+
step 3830000 | 17059s | train_loss=0.0293
|
| 386 |
+
step 3840000 | 17095s | train_loss=0.0280
|
| 387 |
+
step 3850000 | 17132s | train_loss=0.0310
|
| 388 |
+
step 3860000 | 17169s | train_loss=0.0286
|
| 389 |
+
step 3870000 | 17206s | train_loss=0.0279
|
| 390 |
+
step 3880000 | 17243s | train_loss=0.0295
|
| 391 |
+
step 3890000 | 17280s | train_loss=0.0277
|
| 392 |
+
step 3900000 | 17331s | train_loss=0.0288 | fid=nan
|
| 393 |
+
step 3910000 | 17368s | train_loss=0.0294
|
| 394 |
+
step 3920000 | 17405s | train_loss=0.0301
|
| 395 |
+
step 3930000 | 17441s | train_loss=0.0300
|
| 396 |
+
step 3940000 | 17472s | train_loss=0.0288
|
| 397 |
+
step 3950000 | 17509s | train_loss=0.0305
|
| 398 |
+
step 3960000 | 17545s | train_loss=0.0296
|
| 399 |
+
step 3970000 | 17582s | train_loss=0.0318
|
| 400 |
+
step 3980000 | 17619s | train_loss=0.0261
|
| 401 |
+
step 3990000 | 17656s | train_loss=0.0324
|
| 402 |
+
step 4000000 | 17707s | train_loss=0.0272 | fid=nan
|
| 403 |
+
step 4010000 | 17744s | train_loss=0.0273
|
| 404 |
+
step 4020000 | 17781s | train_loss=0.0280
|
| 405 |
+
step 4030000 | 17818s | train_loss=0.0295
|
| 406 |
+
step 4040000 | 17855s | train_loss=0.0288
|
| 407 |
+
step 4050000 | 17887s | train_loss=0.0271
|
| 408 |
+
step 4060000 | 17920s | train_loss=0.0283
|
| 409 |
+
step 4070000 | 17957s | train_loss=0.0268
|
| 410 |
+
step 4080000 | 17994s | train_loss=0.0319
|
| 411 |
+
step 4090000 | 18031s | train_loss=0.0287
|
| 412 |
+
step 4100000 | 18082s | train_loss=0.0285 | fid=nan
|
| 413 |
+
step 4110000 | 18119s | train_loss=0.0280
|
| 414 |
+
step 4120000 | 18156s | train_loss=0.0295
|
| 415 |
+
step 4130000 | 18193s | train_loss=0.0283
|
| 416 |
+
step 4140000 | 18230s | train_loss=0.0289
|
| 417 |
+
step 4150000 | 18266s | train_loss=0.0253
|
| 418 |
+
step 4160000 | 18303s | train_loss=0.0278
|
| 419 |
+
step 4170000 | 18333s | train_loss=0.0280
|
| 420 |
+
step 4180000 | 18370s | train_loss=0.0302
|
| 421 |
+
step 4190000 | 18407s | train_loss=0.0277
|
| 422 |
+
step 4200000 | 18458s | train_loss=0.0287 | fid=nan
|
| 423 |
+
step 4210000 | 18495s | train_loss=0.0282
|
| 424 |
+
step 4220000 | 18532s | train_loss=0.0276
|
| 425 |
+
step 4230000 | 18569s | train_loss=0.0272
|
| 426 |
+
step 4240000 | 18606s | train_loss=0.0261
|
| 427 |
+
step 4250000 | 18643s | train_loss=0.0292
|
| 428 |
+
step 4260000 | 18680s | train_loss=0.0257
|
| 429 |
+
step 4270000 | 18717s | train_loss=0.0265
|
| 430 |
+
step 4280000 | 18749s | train_loss=0.0270
|
| 431 |
+
step 4290000 | 18785s | train_loss=0.0268
|
| 432 |
+
step 4300000 | 18836s | train_loss=0.0271 | fid=nan
|
| 433 |
+
step 4310000 | 18873s | train_loss=0.0263
|
| 434 |
+
step 4320000 | 18910s | train_loss=0.0260
|
| 435 |
+
step 4330000 | 18947s | train_loss=0.0261
|
| 436 |
+
step 4340000 | 18984s | train_loss=0.0305
|
| 437 |
+
step 4350000 | 19021s | train_loss=0.0293
|
| 438 |
+
step 4360000 | 19058s | train_loss=0.0275
|
| 439 |
+
step 4370000 | 19095s | train_loss=0.0289
|
| 440 |
+
step 4380000 | 19132s | train_loss=0.0277
|
| 441 |
+
step 4390000 | 19168s | train_loss=0.0294
|
| 442 |
+
step 4400000 | 19215s | train_loss=0.0269 | fid=nan
|
| 443 |
+
step 4410000 | 19252s | train_loss=0.0275
|
| 444 |
+
step 4420000 | 19289s | train_loss=0.0270
|
| 445 |
+
step 4430000 | 19326s | train_loss=0.0277
|
| 446 |
+
step 4440000 | 19363s | train_loss=0.0265
|
| 447 |
+
step 4450000 | 19400s | train_loss=0.0301
|
| 448 |
+
step 4460000 | 19437s | train_loss=0.0288
|
| 449 |
+
step 4470000 | 19473s | train_loss=0.0258
|
| 450 |
+
step 4480000 | 19510s | train_loss=0.0259
|
| 451 |
+
step 4490000 | 19547s | train_loss=0.0288
|
| 452 |
+
step 4500000 | 19595s | train_loss=0.0289 | fid=nan
|
| 453 |
+
step 4510000 | 19628s | train_loss=0.0268
|
| 454 |
+
step 4520000 | 19665s | train_loss=0.0314
|
| 455 |
+
step 4530000 | 19702s | train_loss=0.0269
|
| 456 |
+
step 4540000 | 19739s | train_loss=0.0264
|
| 457 |
+
step 4550000 | 19776s | train_loss=0.0330
|
| 458 |
+
step 4560000 | 19813s | train_loss=0.0263
|
| 459 |
+
step 4570000 | 19850s | train_loss=0.0269
|
| 460 |
+
step 4580000 | 19887s | train_loss=0.0272
|
| 461 |
+
step 4590000 | 19923s | train_loss=0.0255
|
| 462 |
+
step 4600000 | 19998s | train_loss=0.0298 | fid=156.3187
|
| 463 |
+
step 4610000 | 20030s | train_loss=0.0268
|
| 464 |
+
step 4620000 | 20067s | train_loss=0.0281
|
| 465 |
+
step 4630000 | 20104s | train_loss=0.0285
|
| 466 |
+
step 4640000 | 20141s | train_loss=0.0262
|
| 467 |
+
step 4650000 | 20178s | train_loss=0.0253
|
| 468 |
+
step 4660000 | 20216s | train_loss=0.0287
|
| 469 |
+
step 4670000 | 20253s | train_loss=0.0273
|
| 470 |
+
step 4680000 | 20290s | train_loss=0.0265
|
| 471 |
+
step 4690000 | 20327s | train_loss=0.0264
|
| 472 |
+
step 4700000 | 20401s | train_loss=0.0282 | fid=178.3293
|
| 473 |
+
step 4710000 | 20432s | train_loss=0.0248
|
| 474 |
+
step 4720000 | 20469s | train_loss=0.0242
|
| 475 |
+
step 4730000 | 20506s | train_loss=0.0253
|
| 476 |
+
step 4740000 | 20543s | train_loss=0.0287
|
| 477 |
+
step 4750000 | 20580s | train_loss=0.0265
|
| 478 |
+
step 4760000 | 20617s | train_loss=0.0249
|
| 479 |
+
step 4770000 | 20654s | train_loss=0.0288
|
| 480 |
+
step 4780000 | 20691s | train_loss=0.0261
|
| 481 |
+
step 4790000 | 20727s | train_loss=0.0265
|
| 482 |
+
step 4800000 | 20802s | train_loss=0.0257 | fid=157.1534
|
| 483 |
+
step 4810000 | 20839s | train_loss=0.0281
|
| 484 |
+
step 4820000 | 20868s | train_loss=0.0266
|
| 485 |
+
step 4830000 | 20905s | train_loss=0.0287
|
| 486 |
+
step 4840000 | 20942s | train_loss=0.0259
|
| 487 |
+
step 4850000 | 20979s | train_loss=0.0249
|
| 488 |
+
step 4860000 | 21016s | train_loss=0.0273
|
| 489 |
+
step 4870000 | 21053s | train_loss=0.0244
|
| 490 |
+
step 4880000 | 21090s | train_loss=0.0256
|
| 491 |
+
step 4890000 | 21127s | train_loss=0.0252
|
| 492 |
+
step 4900000 | 21178s | train_loss=0.0269 | fid=nan
|
| 493 |
+
step 4910000 | 21215s | train_loss=0.0248
|
| 494 |
+
step 4920000 | 21251s | train_loss=0.0245
|
| 495 |
+
step 4930000 | 21280s | train_loss=0.0258
|
| 496 |
+
step 4940000 | 21317s | train_loss=0.0268
|
| 497 |
+
step 4950000 | 21354s | train_loss=0.0259
|
| 498 |
+
step 4960000 | 21391s | train_loss=0.0258
|
| 499 |
+
step 4970000 | 21428s | train_loss=0.0251
|
| 500 |
+
step 4980000 | 21465s | train_loss=0.0244
|
| 501 |
+
step 4990000 | 21502s | train_loss=0.0255
|
| 502 |
+
step 5000000 | 21574s | train_loss=0.0261 | fid=176.1552
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d260/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d260/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d260",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 44,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d260/metrics.jsonl
ADDED
|
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"event": "meta", "data": "cifar10_diverse1k", "model": "mlp", "hidden": 1024, "depth": 5, "optimizer": "sgd", "lr": 0.001, "momentum": 0.8, "d_latent": 260, "n": 1000, "vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d260/vae.pt", "subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json", "subset_indices": [41905, 27590, 3837, 45564, 49079, 45615, 24791, 29586, 40154, 14155, 41744, 31565, 21399, 47093, 14144, 7260, 25109, 49111, 3286, 28559, 8088, 18636, 27677, 5584, 14605, 28090, 44911, 22197, 30774, 45976, 21377, 38974, 37419, 34658, 48141, 45353, 16718, 44442, 14512, 30061, 32910, 17321, 15565, 17405, 44045, 17717, 6400, 37998, 3773, 20538, 19316, 34889, 4564, 1363, 5611, 41687, 16986, 41914, 25046, 49823, 33007, 20861, 27988, 42869, 33687, 3828, 12070, 15418, 22140, 35636, 31728, 9480, 7712, 2852, 43042, 17354, 27208, 49689, 14774, 29225, 25390, 35055, 12670, 19522, 24246, 14561, 33656, 42751, 2242, 364, 43176, 15490, 46231, 26061, 18629, 34392, 32488, 26106, 10393, 11167, 20874, 8106, 10093, 7243, 48907, 8147, 45063, 21102, 37655, 7635, 2514, 27684, 3204, 28758, 28311, 29718, 44311, 10355, 32304, 14615, 28486, 14294, 16760, 37166, 19433, 20899, 23979, 36854, 43726, 10375, 11078, 10677, 40127, 33642, 36325, 13174, 19662, 30319, 22501, 14354, 24243, 27654, 42509, 18915, 29668, 2313, 42454, 11218, 48097, 31985, 16567, 26984, 33502, 36329, 22928, 42746, 38223, 19499, 41926, 29059, 44004, 8962, 40807, 11854, 36894, 13566, 36982, 16226, 36410, 47002, 20194, 3625, 19297, 2555, 4842, 8098, 44052, 5795, 5274, 4374, 17924, 30344, 5386, 26317, 30389, 32399, 32042, 26731, 29532, 33817, 26942, 25858, 11887, 46070, 41837, 10713, 33955, 20000, 29946, 4912, 32112, 6697, 44031, 27296, 37785, 25660, 27948, 31260, 24358, 1827, 37445, 25084, 11703, 3820, 49218, 30616, 25083, 22759, 6275, 40603, 39038, 37156, 8894, 9871, 31333, 26582, 20528, 10523, 2747, 33639, 41740, 47899, 11337, 6325, 433, 33051, 1094, 24743, 9392, 22638, 7689, 39641, 9486, 33554, 30840, 35032, 19647, 7903, 18205, 34063, 46483, 32963, 11601, 19203, 45792, 47136, 6174, 40793, 13885, 16641, 48744, 37658, 43829, 44107, 6388, 37424, 20033, 28563, 27013, 17032, 12768, 38181, 35608, 14089, 29228, 40229, 45106, 11963, 37979, 20694, 22835, 5773, 19580, 44596, 19079, 45191, 666, 15893, 7513, 15939, 45020, 12659, 31388, 44651, 23592, 16419, 47615, 36714, 36448, 28579, 1697, 9889, 42082, 49150, 41132, 33996, 21551, 18467, 45242, 6516, 3355, 20187, 31669, 2814, 6900, 28198, 17656, 18178, 31825, 34244, 38330, 28894, 7463, 8134, 38159, 10524, 41508, 2367, 496, 31046, 1511, 31956, 46621, 14176, 1899, 32397, 29452, 16504, 34042, 24014, 47447, 20229, 36683, 5186, 14349, 21736, 17094, 30311, 27949, 30023, 49589, 43270, 2898, 18259, 17175, 18594, 17062, 42063, 3237, 12080, 35769, 1687, 30330, 29678, 27321, 4273, 28513, 2551, 5314, 13572, 13497, 9652, 3466, 27133, 38459, 23232, 14231, 25584, 49398, 32994, 10434, 42369, 37410, 45044, 11075, 12523, 1417, 6765, 40000, 25016, 36905, 48638, 21, 16822, 24333, 30084, 43982, 26954, 6519, 42546, 14751, 14278, 49970, 47621, 18883, 37295, 9040, 40369, 22225, 36489, 21302, 13110, 5455, 47804, 14868, 13063, 22102, 36825, 21969, 33186, 28908, 32592, 42766, 12262, 39073, 915, 11824, 39208, 3098, 42789, 29800, 14150, 28705, 8847, 21716, 535, 48077, 41540, 36207, 10816, 41471, 35049, 20661, 6083, 30194, 4576, 2174, 1908, 48392, 10435, 10986, 18036, 24746, 3941, 34118, 49749, 12025, 21493, 15785, 17085, 39411, 34561, 42064, 36237, 2740, 8183, 40564, 33107, 25765, 49165, 25211, 17520, 17115, 1869, 46432, 33757, 20517, 39760, 259, 45380, 12436, 48000, 21046, 4039, 15701, 35246, 29112, 44161, 34079, 45449, 42845, 22815, 5597, 3990, 35899, 48969, 41528, 27873, 16598, 39104, 23359, 46847, 28624, 19198, 1251, 9003, 28383, 47005, 40703, 41605, 39895, 26201, 11727, 27435, 17333, 39261, 37456, 14653, 30634, 4160, 24540, 12044, 7111, 23485, 5645, 2572, 1682, 17872, 8974, 16796, 11864, 40178, 20092, 32219, 14458, 37007, 31340, 10520, 34489, 13128, 42913, 31468, 44180, 37938, 20058, 18673, 36850, 34721, 26210, 15257, 35188, 11815, 47936, 18088, 43578, 11996, 47952, 24311, 1884, 36557, 32921, 36250, 38022, 5671, 13621, 23326, 14326, 23471, 24685, 39092, 3268, 37257, 16348, 13730, 21838, 44088, 8945, 10827, 13840, 46123, 48010, 13350, 30622, 45233, 8698, 14335, 45842, 15138, 1515, 40759, 21960, 47551, 33208, 18260, 6377, 46869, 31742, 23566, 12829, 15305, 37988, 16111, 32146, 37948, 25917, 30484, 17885, 42468, 28238, 33865, 28030, 30072, 34855, 19324, 42075, 30623, 3667, 38832, 25100, 43077, 36471, 11663, 13588, 14015, 5099, 10589, 28526, 48022, 15444, 3280, 188, 31582, 15225, 11019, 21804, 3901, 24461, 25031, 10373, 4486, 45950, 37949, 32923, 1583, 6576, 26004, 16237, 38132, 25045, 25091, 41949, 11553, 48062, 40503, 13057, 8954, 18566, 28405, 28604, 30434, 49112, 30676, 32930, 16183, 48336, 43162, 873, 37233, 44718, 37247, 23758, 9243, 12126, 5287, 49594, 3459, 32091, 20894, 41743, 7366, 7258, 37939, 47665, 12260, 20836, 43253, 16191, 11619, 1519, 6137, 3934, 48375, 35234, 103, 30510, 8869, 21085, 41234, 25138, 25079, 31218, 48292, 48814, 6635, 18416, 1421, 4363, 5164, 47455, 21280, 16767, 19685, 16449, 45798, 31292, 31049, 43587, 30772, 2651, 12854, 23199, 25234, 23389, 24742, 19480, 19738, 19726, 2388, 21394, 4512, 9069, 42856, 25766, 23797, 40741, 33624, 8328, 35922, 27732, 36268, 23213, 41966, 40002, 12750, 46225, 35182, 23769, 25152, 26764, 20090, 27202, 48096, 4266, 2205, 2020, 37651, 36088, 1021, 28544, 20933, 48924, 11503, 43997, 9492, 19505, 47632, 4118, 2029, 22008, 41396, 16378, 36738, 33194, 28053, 44137, 30771, 12131, 5963, 31754, 46953, 3180, 1718, 36816, 2208, 20898, 13799, 30796, 34445, 37540, 27983, 39216, 16791, 36763, 15987, 28440, 11782, 48531, 4252, 47359, 5351, 4059, 9263, 32274, 39283, 258, 17485, 31280, 12814, 33217, 38203, 11170, 8486, 14434, 13722, 36048, 3705, 47210, 38260, 2200, 17850, 21589, 27598, 9926, 9419, 3981, 34519, 37821, 22855, 39796, 24105, 5113, 5888, 31974, 40040, 17179, 48615, 38665, 47173, 47154, 20619, 5049, 1656, 27736, 14424, 2013, 1348, 32964, 37228, 34968, 34124, 15993, 42420, 26558, 2961, 27119, 37779, 29321, 31595, 29301, 38364, 17528, 19583, 48609, 4428, 12427, 601, 21213, 44456, 39274, 40250, 39804, 35373, 43875, 20550, 43426, 13696, 34572, 11097, 22458, 33042, 12981, 29073, 6601, 2370, 4065, 46317, 29460, 21761, 13977, 4520, 31604, 16259, 1445, 29204, 26097, 31921, 33193, 36054, 26372, 4436, 40306, 13974, 6159, 46233, 33714, 27592, 23153, 40311, 12294, 40141, 14030, 36304, 40086, 43163, 14689, 5002, 40885, 40758, 42458, 18580, 29784, 14527, 30070, 18326, 40830, 42503, 5434, 25179, 15652, 11428, 40980, 45916, 22056, 15896, 42628, 30092, 40834, 25862, 40370, 7804, 46908, 44191, 26908, 9764, 36969, 21521, 35257, 9279, 36964, 4012, 627, 31813, 15669, 33832, 29874, 1912, 41254, 23709, 4648, 3131, 49882, 19349, 35975, 25565, 21641, 48219, 30806, 33979, 36103, 14640, 34524, 31661, 26869, 27297, 46519, 2352, 46930, 7617, 49750, 12620, 29820, 18590, 890, 16394, 23877, 26421, 23417, 29021, 11596, 22017, 14794, 14437, 48385, 48231, 18946, 38048, 41302, 40633, 15317, 24820], "fid_reference": "same farthest-first 1k CIFAR-10 training subset", "seed": 44}
|
| 2 |
+
{"step": 1, "wall_time": 49.78078603744507, "train_loss_step": 258.6740417480469, "n_params": 4797700, "total_flops": 14738534400, "train_loss": 1.0021064171424279, "subset_score_loss": 1.0021064171424279, "memorization_fraction_pixel": 0.0, "mean_nn_ratio_pixel": 0.9826973676681519, "fid": 301.1250180113045, "fid_failed": 0.0}
|
| 3 |
+
{"step": 10000, "wall_time": 109.6675775051117, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 147385344000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 4 |
+
{"step": 20000, "wall_time": 169.39005661010742, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 294770688000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 5 |
+
{"step": 30000, "wall_time": 229.4095962047577, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 442156032000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 6 |
+
{"step": 40000, "wall_time": 289.44079399108887, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 589541376000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 7 |
+
{"step": 50000, "wall_time": 349.48121452331543, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 736926720000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 8 |
+
{"step": 60000, "wall_time": 409.5580289363861, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 884312064000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 9 |
+
{"step": 70000, "wall_time": 469.57281732559204, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 1031697408000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 10 |
+
{"step": 80000, "wall_time": 529.4610652923584, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 1179082752000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 11 |
+
{"step": 90000, "wall_time": 589.4048595428467, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 1326468096000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 12 |
+
{"step": 100000, "wall_time": 666.894424200058, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 1473853440000000, "train_loss": NaN, "subset_score_loss": NaN, "memorization_fraction_pixel": 0.0, "mean_nn_ratio_pixel": NaN, "fid": NaN, "fid_failed": 1.0}
|
| 13 |
+
{"step": 110000, "wall_time": 722.8849613666534, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 1621238784000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 14 |
+
{"step": 120000, "wall_time": 782.7981467247009, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 1768624128000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 15 |
+
{"step": 130000, "wall_time": 842.8615109920502, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 1916009472000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 16 |
+
{"step": 140000, "wall_time": 902.7936663627625, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 2063394816000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 17 |
+
{"step": 150000, "wall_time": 962.5660166740417, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 2210780160000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 18 |
+
{"step": 160000, "wall_time": 1022.528995513916, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 2358165504000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 19 |
+
{"step": 170000, "wall_time": 1082.5894312858582, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 2505550848000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 20 |
+
{"step": 180000, "wall_time": 1142.60795378685, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 2652936192000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 21 |
+
{"step": 190000, "wall_time": 1202.7557611465454, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 2800321536000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 22 |
+
{"step": 200000, "wall_time": 1280.443773508072, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 2947706880000000, "train_loss": NaN, "subset_score_loss": NaN, "memorization_fraction_pixel": 0.0, "mean_nn_ratio_pixel": NaN, "fid": NaN, "fid_failed": 1.0}
|
| 23 |
+
{"step": 210000, "wall_time": 1337.0013201236725, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 3095092224000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 24 |
+
{"step": 220000, "wall_time": 1396.548373222351, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 3242477568000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 25 |
+
{"step": 230000, "wall_time": 1456.556159734726, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 3389862912000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 26 |
+
{"step": 240000, "wall_time": 1516.5888636112213, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 3537248256000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 27 |
+
{"step": 250000, "wall_time": 1576.635136604309, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 3684633600000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 28 |
+
{"step": 260000, "wall_time": 1636.8602101802826, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 3832018944000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 29 |
+
{"step": 270000, "wall_time": 1696.8633923530579, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 3979404288000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 30 |
+
{"step": 280000, "wall_time": 1756.9318017959595, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 4126789632000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 31 |
+
{"step": 290000, "wall_time": 1816.8977346420288, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 4274174976000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 32 |
+
{"step": 300000, "wall_time": 1894.4660396575928, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 4421560320000000, "train_loss": NaN, "subset_score_loss": NaN, "memorization_fraction_pixel": 0.0, "mean_nn_ratio_pixel": NaN, "fid": NaN, "fid_failed": 1.0}
|
| 33 |
+
{"step": 310000, "wall_time": 1955.1683423519135, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 4568945664000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 34 |
+
{"step": 320000, "wall_time": 2010.9355878829956, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 4716331008000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 35 |
+
{"step": 330000, "wall_time": 2070.8666808605194, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 4863716352000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 36 |
+
{"step": 340000, "wall_time": 2130.918090581894, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 5011101696000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 37 |
+
{"step": 350000, "wall_time": 2190.7500829696655, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 5158487040000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 38 |
+
{"step": 360000, "wall_time": 2250.838714361191, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 5305872384000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 39 |
+
{"step": 370000, "wall_time": 2311.033217906952, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 5453257728000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 40 |
+
{"step": 380000, "wall_time": 2371.1494994163513, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 5600643072000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 41 |
+
{"step": 390000, "wall_time": 2431.0478298664093, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 5748028416000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 42 |
+
{"step": 400000, "wall_time": 2509.99063706398, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 5895413760000000, "train_loss": NaN, "subset_score_loss": NaN, "memorization_fraction_pixel": 0.0, "mean_nn_ratio_pixel": NaN, "fid": NaN, "fid_failed": 1.0}
|
| 43 |
+
{"step": 410000, "wall_time": 2570.080740213394, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 6042799104000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 44 |
+
{"step": 420000, "wall_time": 2626.784148454666, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 6190184448000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 45 |
+
{"step": 430000, "wall_time": 2686.76033616066, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 6337569792000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 46 |
+
{"step": 440000, "wall_time": 2746.680551290512, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 6484955136000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 47 |
+
{"step": 450000, "wall_time": 2806.5258872509003, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 6632340480000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 48 |
+
{"step": 460000, "wall_time": 2866.5420928001404, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 6779725824000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 49 |
+
{"step": 470000, "wall_time": 2926.584854364395, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 6927111168000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 50 |
+
{"step": 480000, "wall_time": 2986.7105193138123, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 7074496512000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 51 |
+
{"step": 490000, "wall_time": 3046.8727474212646, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 7221881856000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 52 |
+
{"step": 500000, "wall_time": 3125.495137691498, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 7369267200000000, "train_loss": NaN, "subset_score_loss": NaN, "memorization_fraction_pixel": 0.0, "mean_nn_ratio_pixel": NaN, "fid": NaN, "fid_failed": 1.0}
|
| 53 |
+
{"step": 510000, "wall_time": 3185.3527569770813, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 7516652544000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 54 |
+
{"step": 520000, "wall_time": 3242.7607769966125, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 7664037888000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 55 |
+
{"step": 530000, "wall_time": 3302.690354824066, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 7811423232000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 56 |
+
{"step": 540000, "wall_time": 3362.7475843429565, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 7958808576000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 57 |
+
{"step": 550000, "wall_time": 3422.7730038166046, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 8106193920000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 58 |
+
{"step": 560000, "wall_time": 3482.990183353424, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 8253579264000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 59 |
+
{"step": 570000, "wall_time": 3543.0520827770233, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 8400964608000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 60 |
+
{"step": 580000, "wall_time": 3603.29531621933, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 8548349952000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 61 |
+
{"step": 590000, "wall_time": 3663.136749982834, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 8695735296000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 62 |
+
{"step": 600000, "wall_time": 3741.98557138443, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 8843120640000000, "train_loss": NaN, "subset_score_loss": NaN, "memorization_fraction_pixel": 0.0, "mean_nn_ratio_pixel": NaN, "fid": NaN, "fid_failed": 1.0}
|
| 63 |
+
{"step": 610000, "wall_time": 3801.997993707657, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 8990505984000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 64 |
+
{"step": 620000, "wall_time": 3862.473399877548, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 9137891328000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 65 |
+
{"step": 630000, "wall_time": 3918.6428096294403, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 9285276672000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 66 |
+
{"step": 640000, "wall_time": 3978.708647251129, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 9432662016000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 67 |
+
{"step": 650000, "wall_time": 4039.0649724006653, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 9580047360000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 68 |
+
{"step": 660000, "wall_time": 4099.120117902756, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 9727432704000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 69 |
+
{"step": 670000, "wall_time": 4159.133900642395, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 9874818048000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 70 |
+
{"step": 680000, "wall_time": 4219.10445690155, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 10022203392000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 71 |
+
{"step": 690000, "wall_time": 4275.866033554077, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 10169588736000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 72 |
+
{"step": 700000, "wall_time": 4357.5216274261475, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 10316974080000000, "train_loss": NaN, "subset_score_loss": NaN, "memorization_fraction_pixel": 0.0, "mean_nn_ratio_pixel": NaN, "fid": NaN, "fid_failed": 1.0}
|
| 73 |
+
{"step": 710000, "wall_time": 4417.596939325333, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 10464359424000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 74 |
+
{"step": 720000, "wall_time": 4477.6698434352875, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 10611744768000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 75 |
+
{"step": 730000, "wall_time": 4534.162224292755, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 10759130112000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 76 |
+
{"step": 740000, "wall_time": 4594.311327219009, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 10906515456000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 77 |
+
{"step": 750000, "wall_time": 4654.127651691437, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 11053900800000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 78 |
+
{"step": 760000, "wall_time": 4714.118191719055, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 11201286144000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 79 |
+
{"step": 770000, "wall_time": 4774.218793869019, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 11348671488000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 80 |
+
{"step": 780000, "wall_time": 4834.289823770523, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 11496056832000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 81 |
+
{"step": 790000, "wall_time": 4889.212495088577, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 11643442176000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 82 |
+
{"step": 800000, "wall_time": 4972.567769765854, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 11790827520000000, "train_loss": NaN, "subset_score_loss": NaN, "memorization_fraction_pixel": 0.0, "mean_nn_ratio_pixel": NaN, "fid": NaN, "fid_failed": 1.0}
|
| 83 |
+
{"step": 810000, "wall_time": 5032.519511938095, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 11938212864000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 84 |
+
{"step": 820000, "wall_time": 5092.599521160126, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 12085598208000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 85 |
+
{"step": 830000, "wall_time": 5151.926220417023, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 12232983552000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 86 |
+
{"step": 840000, "wall_time": 5210.191227197647, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 12380368896000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 87 |
+
{"step": 850000, "wall_time": 5270.484299898148, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 12527754240000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 88 |
+
{"step": 860000, "wall_time": 5330.38162279129, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 12675139584000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 89 |
+
{"step": 870000, "wall_time": 5390.381986856461, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 12822524928000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 90 |
+
{"step": 880000, "wall_time": 5450.4160578250885, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 12969910272000000, "train_loss": NaN, "subset_score_loss": NaN}
|
| 91 |
+
{"step": 890000, "wall_time": 5505.362161397934, "train_loss_step": NaN, "n_params": 4797700, "total_flops": 13117295616000000, "train_loss": NaN, "subset_score_loss": NaN}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d260/runner.log
ADDED
|
@@ -0,0 +1,91 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 50s | train_loss=1.0021 | fid=301.1250
|
| 3 |
+
step 10000 | 110s | train_loss=nan
|
| 4 |
+
step 20000 | 169s | train_loss=nan
|
| 5 |
+
step 30000 | 229s | train_loss=nan
|
| 6 |
+
step 40000 | 289s | train_loss=nan
|
| 7 |
+
step 50000 | 349s | train_loss=nan
|
| 8 |
+
step 60000 | 410s | train_loss=nan
|
| 9 |
+
step 70000 | 470s | train_loss=nan
|
| 10 |
+
step 80000 | 529s | train_loss=nan
|
| 11 |
+
step 90000 | 589s | train_loss=nan
|
| 12 |
+
step 100000 | 667s | train_loss=nan | fid=nan
|
| 13 |
+
step 110000 | 723s | train_loss=nan
|
| 14 |
+
step 120000 | 783s | train_loss=nan
|
| 15 |
+
step 130000 | 843s | train_loss=nan
|
| 16 |
+
step 140000 | 903s | train_loss=nan
|
| 17 |
+
step 150000 | 963s | train_loss=nan
|
| 18 |
+
step 160000 | 1023s | train_loss=nan
|
| 19 |
+
step 170000 | 1083s | train_loss=nan
|
| 20 |
+
step 180000 | 1143s | train_loss=nan
|
| 21 |
+
step 190000 | 1203s | train_loss=nan
|
| 22 |
+
step 200000 | 1280s | train_loss=nan | fid=nan
|
| 23 |
+
step 210000 | 1337s | train_loss=nan
|
| 24 |
+
step 220000 | 1397s | train_loss=nan
|
| 25 |
+
step 230000 | 1457s | train_loss=nan
|
| 26 |
+
step 240000 | 1517s | train_loss=nan
|
| 27 |
+
step 250000 | 1577s | train_loss=nan
|
| 28 |
+
step 260000 | 1637s | train_loss=nan
|
| 29 |
+
step 270000 | 1697s | train_loss=nan
|
| 30 |
+
step 280000 | 1757s | train_loss=nan
|
| 31 |
+
step 290000 | 1817s | train_loss=nan
|
| 32 |
+
step 300000 | 1894s | train_loss=nan | fid=nan
|
| 33 |
+
step 310000 | 1955s | train_loss=nan
|
| 34 |
+
step 320000 | 2011s | train_loss=nan
|
| 35 |
+
step 330000 | 2071s | train_loss=nan
|
| 36 |
+
step 340000 | 2131s | train_loss=nan
|
| 37 |
+
step 350000 | 2191s | train_loss=nan
|
| 38 |
+
step 360000 | 2251s | train_loss=nan
|
| 39 |
+
step 370000 | 2311s | train_loss=nan
|
| 40 |
+
step 380000 | 2371s | train_loss=nan
|
| 41 |
+
step 390000 | 2431s | train_loss=nan
|
| 42 |
+
step 400000 | 2510s | train_loss=nan | fid=nan
|
| 43 |
+
step 410000 | 2570s | train_loss=nan
|
| 44 |
+
step 420000 | 2627s | train_loss=nan
|
| 45 |
+
step 430000 | 2687s | train_loss=nan
|
| 46 |
+
step 440000 | 2747s | train_loss=nan
|
| 47 |
+
step 450000 | 2807s | train_loss=nan
|
| 48 |
+
step 460000 | 2867s | train_loss=nan
|
| 49 |
+
step 470000 | 2927s | train_loss=nan
|
| 50 |
+
step 480000 | 2987s | train_loss=nan
|
| 51 |
+
step 490000 | 3047s | train_loss=nan
|
| 52 |
+
step 500000 | 3125s | train_loss=nan | fid=nan
|
| 53 |
+
step 510000 | 3185s | train_loss=nan
|
| 54 |
+
step 520000 | 3243s | train_loss=nan
|
| 55 |
+
step 530000 | 3303s | train_loss=nan
|
| 56 |
+
step 540000 | 3363s | train_loss=nan
|
| 57 |
+
step 550000 | 3423s | train_loss=nan
|
| 58 |
+
step 560000 | 3483s | train_loss=nan
|
| 59 |
+
step 570000 | 3543s | train_loss=nan
|
| 60 |
+
step 580000 | 3603s | train_loss=nan
|
| 61 |
+
step 590000 | 3663s | train_loss=nan
|
| 62 |
+
step 600000 | 3742s | train_loss=nan | fid=nan
|
| 63 |
+
step 610000 | 3802s | train_loss=nan
|
| 64 |
+
step 620000 | 3862s | train_loss=nan
|
| 65 |
+
step 630000 | 3919s | train_loss=nan
|
| 66 |
+
step 640000 | 3979s | train_loss=nan
|
| 67 |
+
step 650000 | 4039s | train_loss=nan
|
| 68 |
+
step 660000 | 4099s | train_loss=nan
|
| 69 |
+
step 670000 | 4159s | train_loss=nan
|
| 70 |
+
step 680000 | 4219s | train_loss=nan
|
| 71 |
+
step 690000 | 4276s | train_loss=nan
|
| 72 |
+
step 700000 | 4358s | train_loss=nan | fid=nan
|
| 73 |
+
step 710000 | 4418s | train_loss=nan
|
| 74 |
+
step 720000 | 4478s | train_loss=nan
|
| 75 |
+
step 730000 | 4534s | train_loss=nan
|
| 76 |
+
step 740000 | 4594s | train_loss=nan
|
| 77 |
+
step 750000 | 4654s | train_loss=nan
|
| 78 |
+
step 760000 | 4714s | train_loss=nan
|
| 79 |
+
step 770000 | 4774s | train_loss=nan
|
| 80 |
+
step 780000 | 4834s | train_loss=nan
|
| 81 |
+
step 790000 | 4889s | train_loss=nan
|
| 82 |
+
step 800000 | 4973s | train_loss=nan | fid=nan
|
| 83 |
+
step 810000 | 5033s | train_loss=nan
|
| 84 |
+
step 820000 | 5093s | train_loss=nan
|
| 85 |
+
step 830000 | 5152s | train_loss=nan
|
| 86 |
+
step 840000 | 5210s | train_loss=nan
|
| 87 |
+
step 850000 | 5270s | train_loss=nan
|
| 88 |
+
step 860000 | 5330s | train_loss=nan
|
| 89 |
+
step 870000 | 5390s | train_loss=nan
|
| 90 |
+
step 880000 | 5450s | train_loss=nan
|
| 91 |
+
step 890000 | 5505s | train_loss=nan
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d40/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d40/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d40",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 44,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d40/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d40/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 70s | train_loss=0.9936 | fid=229.3860
|
| 3 |
+
step 10000 | 153s | train_loss=0.8923
|
| 4 |
+
step 20000 | 239s | train_loss=0.8683
|
| 5 |
+
step 30000 | 326s | train_loss=0.8120
|
| 6 |
+
step 40000 | 413s | train_loss=0.7019
|
| 7 |
+
step 50000 | 500s | train_loss=0.5223
|
| 8 |
+
step 60000 | 587s | train_loss=0.3511
|
| 9 |
+
step 70000 | 674s | train_loss=0.2476
|
| 10 |
+
step 80000 | 761s | train_loss=0.1966
|
| 11 |
+
step 90000 | 848s | train_loss=0.1652
|
| 12 |
+
step 100000 | 1007s | train_loss=0.1491 | fid=160.5553
|
| 13 |
+
step 110000 | 1089s | train_loss=0.1363
|
| 14 |
+
step 120000 | 1174s | train_loss=0.1257
|
| 15 |
+
step 130000 | 1261s | train_loss=0.1175
|
| 16 |
+
step 140000 | 1348s | train_loss=0.1108
|
| 17 |
+
step 150000 | 1434s | train_loss=0.1037
|
| 18 |
+
step 160000 | 1521s | train_loss=0.0992
|
| 19 |
+
step 170000 | 1608s | train_loss=0.0941
|
| 20 |
+
step 180000 | 1695s | train_loss=0.0892
|
| 21 |
+
step 190000 | 1782s | train_loss=0.0864
|
| 22 |
+
step 200000 | 1941s | train_loss=0.0826 | fid=156.2262
|
| 23 |
+
step 210000 | 2025s | train_loss=0.0802
|
| 24 |
+
step 220000 | 2109s | train_loss=0.0779
|
| 25 |
+
step 230000 | 2194s | train_loss=0.0744
|
| 26 |
+
step 240000 | 2281s | train_loss=0.0723
|
| 27 |
+
step 250000 | 2368s | train_loss=0.0712
|
| 28 |
+
step 260000 | 2455s | train_loss=0.0685
|
| 29 |
+
step 270000 | 2541s | train_loss=0.0653
|
| 30 |
+
step 280000 | 2628s | train_loss=0.0651
|
| 31 |
+
step 290000 | 2715s | train_loss=0.0636
|
| 32 |
+
step 300000 | 2874s | train_loss=0.0626 | fid=152.6525
|
| 33 |
+
step 310000 | 2958s | train_loss=0.0605
|
| 34 |
+
step 320000 | 3046s | train_loss=0.0591
|
| 35 |
+
step 330000 | 3130s | train_loss=0.0583
|
| 36 |
+
step 340000 | 3215s | train_loss=0.0562
|
| 37 |
+
step 350000 | 3302s | train_loss=0.0560
|
| 38 |
+
step 360000 | 3389s | train_loss=0.0546
|
| 39 |
+
step 370000 | 3475s | train_loss=0.0540
|
| 40 |
+
step 380000 | 3562s | train_loss=0.0525
|
| 41 |
+
step 390000 | 3649s | train_loss=0.0521
|
| 42 |
+
step 400000 | 3808s | train_loss=0.0511 | fid=148.5358
|
| 43 |
+
step 410000 | 3892s | train_loss=0.0507
|
| 44 |
+
step 420000 | 3979s | train_loss=0.0490
|
| 45 |
+
step 430000 | 4065s | train_loss=0.0488
|
| 46 |
+
step 440000 | 4151s | train_loss=0.0471
|
| 47 |
+
step 450000 | 4236s | train_loss=0.0473
|
| 48 |
+
step 460000 | 4323s | train_loss=0.0461
|
| 49 |
+
step 470000 | 4410s | train_loss=0.0459
|
| 50 |
+
step 480000 | 4497s | train_loss=0.0454
|
| 51 |
+
step 490000 | 4583s | train_loss=0.0452
|
| 52 |
+
step 500000 | 4740s | train_loss=0.0444 | fid=145.9661
|
| 53 |
+
step 510000 | 4824s | train_loss=0.0437
|
| 54 |
+
step 520000 | 4911s | train_loss=0.0433
|
| 55 |
+
step 530000 | 4998s | train_loss=0.0428
|
| 56 |
+
step 540000 | 5084s | train_loss=0.0426
|
| 57 |
+
step 550000 | 5169s | train_loss=0.0415
|
| 58 |
+
step 560000 | 5256s | train_loss=0.0413
|
| 59 |
+
step 570000 | 5341s | train_loss=0.0419
|
| 60 |
+
step 580000 | 5428s | train_loss=0.0411
|
| 61 |
+
step 590000 | 5515s | train_loss=0.0403
|
| 62 |
+
step 600000 | 5675s | train_loss=0.0403 | fid=144.5259
|
| 63 |
+
step 610000 | 5759s | train_loss=0.0407
|
| 64 |
+
step 620000 | 5845s | train_loss=0.0397
|
| 65 |
+
step 630000 | 5932s | train_loss=0.0382
|
| 66 |
+
step 640000 | 6019s | train_loss=0.0391
|
| 67 |
+
step 650000 | 6104s | train_loss=0.0376
|
| 68 |
+
step 660000 | 6191s | train_loss=0.0378
|
| 69 |
+
step 670000 | 6278s | train_loss=0.0368
|
| 70 |
+
step 680000 | 6363s | train_loss=0.0376
|
| 71 |
+
step 690000 | 6450s | train_loss=0.0366
|
| 72 |
+
step 700000 | 6609s | train_loss=0.0367 | fid=144.9893
|
| 73 |
+
step 710000 | 6693s | train_loss=0.0356
|
| 74 |
+
step 720000 | 6780s | train_loss=0.0362
|
| 75 |
+
step 730000 | 6867s | train_loss=0.0360
|
| 76 |
+
step 740000 | 6953s | train_loss=0.0355
|
| 77 |
+
step 750000 | 7041s | train_loss=0.0347
|
| 78 |
+
step 760000 | 7124s | train_loss=0.0349
|
| 79 |
+
step 770000 | 7211s | train_loss=0.0355
|
| 80 |
+
step 780000 | 7298s | train_loss=0.0344
|
| 81 |
+
step 790000 | 7384s | train_loss=0.0344
|
| 82 |
+
step 800000 | 7543s | train_loss=0.0341 | fid=143.8669
|
| 83 |
+
step 810000 | 7627s | train_loss=0.0338
|
| 84 |
+
step 820000 | 7714s | train_loss=0.0336
|
| 85 |
+
step 830000 | 7801s | train_loss=0.0334
|
| 86 |
+
step 840000 | 7887s | train_loss=0.0327
|
| 87 |
+
step 850000 | 7974s | train_loss=0.0327
|
| 88 |
+
step 860000 | 8060s | train_loss=0.0334
|
| 89 |
+
step 870000 | 8145s | train_loss=0.0320
|
| 90 |
+
step 880000 | 8232s | train_loss=0.0314
|
| 91 |
+
step 890000 | 8319s | train_loss=0.0324
|
| 92 |
+
step 900000 | 8477s | train_loss=0.0314 | fid=142.4444
|
| 93 |
+
step 910000 | 8562s | train_loss=0.0307
|
| 94 |
+
step 920000 | 8649s | train_loss=0.0309
|
| 95 |
+
step 930000 | 8735s | train_loss=0.0313
|
| 96 |
+
step 940000 | 8822s | train_loss=0.0311
|
| 97 |
+
step 950000 | 8909s | train_loss=0.0304
|
| 98 |
+
step 960000 | 8996s | train_loss=0.0303
|
| 99 |
+
step 970000 | 9081s | train_loss=0.0309
|
| 100 |
+
step 980000 | 9167s | train_loss=0.0293
|
| 101 |
+
step 990000 | 9253s | train_loss=0.0294
|
| 102 |
+
step 1000000 | 9413s | train_loss=0.0301 | fid=143.4092
|
| 103 |
+
step 1010000 | 9495s | train_loss=0.0288
|
| 104 |
+
step 1020000 | 9582s | train_loss=0.0294
|
| 105 |
+
step 1030000 | 9669s | train_loss=0.0297
|
| 106 |
+
step 1040000 | 9756s | train_loss=0.0295
|
| 107 |
+
step 1050000 | 9843s | train_loss=0.0283
|
| 108 |
+
step 1060000 | 9929s | train_loss=0.0292
|
| 109 |
+
step 1070000 | 10017s | train_loss=0.0295
|
| 110 |
+
step 1080000 | 10101s | train_loss=0.0285
|
| 111 |
+
step 1090000 | 10186s | train_loss=0.0285
|
| 112 |
+
step 1100000 | 10346s | train_loss=0.0284 | fid=141.8980
|
| 113 |
+
step 1110000 | 10432s | train_loss=0.0283
|
| 114 |
+
step 1120000 | 10515s | train_loss=0.0276
|
| 115 |
+
step 1130000 | 10602s | train_loss=0.0275
|
| 116 |
+
step 1140000 | 10689s | train_loss=0.0278
|
| 117 |
+
step 1150000 | 10775s | train_loss=0.0274
|
| 118 |
+
step 1160000 | 10862s | train_loss=0.0272
|
| 119 |
+
step 1170000 | 10949s | train_loss=0.0271
|
| 120 |
+
step 1180000 | 11036s | train_loss=0.0265
|
| 121 |
+
step 1190000 | 11119s | train_loss=0.0274
|
| 122 |
+
step 1200000 | 11278s | train_loss=0.0271 | fid=142.6714
|
| 123 |
+
step 1210000 | 11365s | train_loss=0.0263
|
| 124 |
+
step 1220000 | 11440s | train_loss=0.0264
|
| 125 |
+
step 1230000 | 11525s | train_loss=0.0271
|
| 126 |
+
step 1240000 | 11611s | train_loss=0.0259
|
| 127 |
+
step 1250000 | 11698s | train_loss=0.0261
|
| 128 |
+
step 1260000 | 11785s | train_loss=0.0266
|
| 129 |
+
step 1270000 | 11872s | train_loss=0.0262
|
| 130 |
+
step 1280000 | 11959s | train_loss=0.0255
|
| 131 |
+
step 1290000 | 12045s | train_loss=0.0258
|
| 132 |
+
step 1300000 | 12201s | train_loss=0.0257 | fid=141.4050
|
| 133 |
+
step 1310000 | 12288s | train_loss=0.0255
|
| 134 |
+
step 1320000 | 12372s | train_loss=0.0251
|
| 135 |
+
step 1330000 | 12459s | train_loss=0.0251
|
| 136 |
+
step 1340000 | 12544s | train_loss=0.0255
|
| 137 |
+
step 1350000 | 12631s | train_loss=0.0247
|
| 138 |
+
step 1360000 | 12717s | train_loss=0.0247
|
| 139 |
+
step 1370000 | 12804s | train_loss=0.0245
|
| 140 |
+
step 1380000 | 12891s | train_loss=0.0248
|
| 141 |
+
step 1390000 | 12978s | train_loss=0.0244
|
| 142 |
+
step 1400000 | 13189s | train_loss=0.0248 | fid=141.5306
|
| 143 |
+
step 1410000 | 13275s | train_loss=0.0244
|
| 144 |
+
step 1420000 | 13349s | train_loss=0.0242
|
| 145 |
+
step 1430000 | 13436s | train_loss=0.0235
|
| 146 |
+
step 1440000 | 13523s | train_loss=0.0236
|
| 147 |
+
step 1450000 | 13608s | train_loss=0.0243
|
| 148 |
+
step 1460000 | 13695s | train_loss=0.0245
|
| 149 |
+
step 1470000 | 13782s | train_loss=0.0237
|
| 150 |
+
step 1480000 | 13869s | train_loss=0.0234
|
| 151 |
+
step 1490000 | 13955s | train_loss=0.0244
|
| 152 |
+
step 1500000 | 14127s | train_loss=0.0229 | fid=141.6092
|
| 153 |
+
step 1510000 | 14213s | train_loss=0.0234
|
| 154 |
+
step 1520000 | 14298s | train_loss=0.0232
|
| 155 |
+
step 1530000 | 14384s | train_loss=0.0238
|
| 156 |
+
step 1540000 | 14471s | train_loss=0.0237
|
| 157 |
+
step 1550000 | 14558s | train_loss=0.0226
|
| 158 |
+
step 1560000 | 14643s | train_loss=0.0232
|
| 159 |
+
step 1570000 | 14730s | train_loss=0.0232
|
| 160 |
+
step 1580000 | 14817s | train_loss=0.0233
|
| 161 |
+
step 1590000 | 14904s | train_loss=0.0225
|
| 162 |
+
step 1600000 | 15062s | train_loss=0.0227 | fid=140.7826
|
| 163 |
+
step 1610000 | 15147s | train_loss=0.0224
|
| 164 |
+
step 1620000 | 15232s | train_loss=0.0229
|
| 165 |
+
step 1630000 | 15319s | train_loss=0.0222
|
| 166 |
+
step 1640000 | 15406s | train_loss=0.0226
|
| 167 |
+
step 1650000 | 15492s | train_loss=0.0221
|
| 168 |
+
step 1660000 | 15579s | train_loss=0.0226
|
| 169 |
+
step 1670000 | 15665s | train_loss=0.0222
|
| 170 |
+
step 1680000 | 15752s | train_loss=0.0213
|
| 171 |
+
step 1690000 | 15838s | train_loss=0.0221
|
| 172 |
+
step 1700000 | 15995s | train_loss=0.0213 | fid=141.4389
|
| 173 |
+
step 1710000 | 16079s | train_loss=0.0219
|
| 174 |
+
step 1720000 | 16157s | train_loss=0.0217
|
| 175 |
+
step 1730000 | 16243s | train_loss=0.0217
|
| 176 |
+
step 1740000 | 16330s | train_loss=0.0216
|
| 177 |
+
step 1750000 | 16416s | train_loss=0.0220
|
| 178 |
+
step 1760000 | 16503s | train_loss=0.0214
|
| 179 |
+
step 1770000 | 16590s | train_loss=0.0220
|
| 180 |
+
step 1780000 | 16675s | train_loss=0.0218
|
| 181 |
+
step 1790000 | 16762s | train_loss=0.0212
|
| 182 |
+
step 1800000 | 16971s | train_loss=0.0212 | fid=140.8917
|
| 183 |
+
step 1810000 | 17053s | train_loss=0.0211
|
| 184 |
+
step 1820000 | 17133s | train_loss=0.0212
|
| 185 |
+
step 1830000 | 17220s | train_loss=0.0211
|
| 186 |
+
step 1840000 | 17307s | train_loss=0.0203
|
| 187 |
+
step 1850000 | 17394s | train_loss=0.0212
|
| 188 |
+
step 1860000 | 17481s | train_loss=0.0206
|
| 189 |
+
step 1870000 | 17568s | train_loss=0.0205
|
| 190 |
+
step 1880000 | 17655s | train_loss=0.0200
|
| 191 |
+
step 1890000 | 17740s | train_loss=0.0213
|
| 192 |
+
step 1900000 | 17900s | train_loss=0.0203 | fid=140.8810
|
| 193 |
+
step 1910000 | 17987s | train_loss=0.0203
|
| 194 |
+
step 1920000 | 18067s | train_loss=0.0208
|
| 195 |
+
step 1930000 | 18154s | train_loss=0.0213
|
| 196 |
+
step 1940000 | 18241s | train_loss=0.0197
|
| 197 |
+
step 1950000 | 18327s | train_loss=0.0199
|
| 198 |
+
step 1960000 | 18414s | train_loss=0.0201
|
| 199 |
+
step 1970000 | 18501s | train_loss=0.0196
|
| 200 |
+
step 1980000 | 18588s | train_loss=0.0206
|
| 201 |
+
step 1990000 | 18675s | train_loss=0.0203
|
| 202 |
+
step 2000000 | 18831s | train_loss=0.0199 | fid=140.8291
|
| 203 |
+
step 2010000 | 18917s | train_loss=0.0192
|
| 204 |
+
step 2020000 | 19002s | train_loss=0.0199
|
| 205 |
+
step 2030000 | 19079s | train_loss=0.0206
|
| 206 |
+
step 2040000 | 19161s | train_loss=0.0194
|
| 207 |
+
step 2050000 | 19248s | train_loss=0.0198
|
| 208 |
+
step 2060000 | 19335s | train_loss=0.0197
|
| 209 |
+
step 2070000 | 19422s | train_loss=0.0196
|
| 210 |
+
step 2080000 | 19509s | train_loss=0.0203
|
| 211 |
+
step 2090000 | 19596s | train_loss=0.0191
|
| 212 |
+
step 2100000 | 19797s | train_loss=0.0190 | fid=141.2786
|
| 213 |
+
step 2110000 | 19883s | train_loss=0.0200
|
| 214 |
+
step 2120000 | 19957s | train_loss=0.0199
|
| 215 |
+
step 2130000 | 20044s | train_loss=0.0187
|
| 216 |
+
step 2140000 | 20121s | train_loss=0.0192
|
| 217 |
+
step 2150000 | 20207s | train_loss=0.0186
|
| 218 |
+
step 2160000 | 20294s | train_loss=0.0185
|
| 219 |
+
step 2170000 | 20381s | train_loss=0.0199
|
| 220 |
+
step 2180000 | 20468s | train_loss=0.0194
|
| 221 |
+
step 2190000 | 20555s | train_loss=0.0194
|
| 222 |
+
step 2200000 | 20767s | train_loss=0.0184 | fid=140.6769
|
| 223 |
+
step 2210000 | 20853s | train_loss=0.0189
|
| 224 |
+
step 2220000 | 20923s | train_loss=0.0187
|
| 225 |
+
step 2230000 | 21010s | train_loss=0.0183
|
| 226 |
+
step 2240000 | 21090s | train_loss=0.0183
|
| 227 |
+
step 2250000 | 21169s | train_loss=0.0182
|
| 228 |
+
step 2260000 | 21256s | train_loss=0.0183
|
| 229 |
+
step 2270000 | 21343s | train_loss=0.0186
|
| 230 |
+
step 2280000 | 21430s | train_loss=0.0183
|
| 231 |
+
step 2290000 | 21517s | train_loss=0.0182
|
| 232 |
+
step 2300000 | 21730s | train_loss=0.0181 | fid=141.4987
|
| 233 |
+
step 2310000 | 21815s | train_loss=0.0178
|
| 234 |
+
step 2320000 | 21892s | train_loss=0.0179
|
| 235 |
+
step 2330000 | 21979s | train_loss=0.0177
|
| 236 |
+
step 2340000 | 22064s | train_loss=0.0185
|
| 237 |
+
step 2350000 | 22139s | train_loss=0.0180
|
| 238 |
+
step 2360000 | 22225s | train_loss=0.0176
|
| 239 |
+
step 2370000 | 22311s | train_loss=0.0179
|
| 240 |
+
step 2380000 | 22398s | train_loss=0.0180
|
| 241 |
+
step 2390000 | 22485s | train_loss=0.0181
|
| 242 |
+
step 2400000 | 22684s | train_loss=0.0183 | fid=141.3747
|
| 243 |
+
step 2410000 | 22769s | train_loss=0.0175
|
| 244 |
+
step 2420000 | 22842s | train_loss=0.0175
|
| 245 |
+
step 2430000 | 22929s | train_loss=0.0177
|
| 246 |
+
step 2440000 | 23015s | train_loss=0.0172
|
| 247 |
+
step 2450000 | 23102s | train_loss=0.0172
|
| 248 |
+
step 2460000 | 23178s | train_loss=0.0178
|
| 249 |
+
step 2470000 | 23264s | train_loss=0.0171
|
| 250 |
+
step 2480000 | 23351s | train_loss=0.0170
|
| 251 |
+
step 2490000 | 23438s | train_loss=0.0175
|
| 252 |
+
step 2500000 | 23653s | train_loss=0.0177 | fid=141.3388
|
| 253 |
+
step 2510000 | 23739s | train_loss=0.0166
|
| 254 |
+
step 2520000 | 23813s | train_loss=0.0168
|
| 255 |
+
step 2530000 | 23900s | train_loss=0.0167
|
| 256 |
+
step 2540000 | 23985s | train_loss=0.0164
|
| 257 |
+
step 2550000 | 24072s | train_loss=0.0170
|
| 258 |
+
step 2560000 | 24153s | train_loss=0.0172
|
| 259 |
+
step 2570000 | 24229s | train_loss=0.0169
|
| 260 |
+
step 2580000 | 24316s | train_loss=0.0163
|
| 261 |
+
step 2590000 | 24402s | train_loss=0.0177
|
| 262 |
+
step 2600000 | 24599s | train_loss=0.0174 | fid=140.5738
|
| 263 |
+
step 2610000 | 24686s | train_loss=0.0176
|
| 264 |
+
step 2620000 | 24761s | train_loss=0.0163
|
| 265 |
+
step 2630000 | 24846s | train_loss=0.0163
|
| 266 |
+
step 2640000 | 24931s | train_loss=0.0175
|
| 267 |
+
step 2650000 | 25018s | train_loss=0.0167
|
| 268 |
+
step 2660000 | 25105s | train_loss=0.0165
|
| 269 |
+
step 2670000 | 25188s | train_loss=0.0169
|
| 270 |
+
step 2680000 | 25265s | train_loss=0.0164
|
| 271 |
+
step 2690000 | 25352s | train_loss=0.0161
|
| 272 |
+
step 2700000 | 25559s | train_loss=0.0162 | fid=141.5225
|
| 273 |
+
step 2710000 | 25646s | train_loss=0.0169
|
| 274 |
+
step 2720000 | 25721s | train_loss=0.0165
|
| 275 |
+
step 2730000 | 25806s | train_loss=0.0162
|
| 276 |
+
step 2740000 | 25892s | train_loss=0.0162
|
| 277 |
+
step 2750000 | 25979s | train_loss=0.0157
|
| 278 |
+
step 2760000 | 26065s | train_loss=0.0163
|
| 279 |
+
step 2770000 | 26152s | train_loss=0.0155
|
| 280 |
+
step 2780000 | 26231s | train_loss=0.0166
|
| 281 |
+
step 2790000 | 26311s | train_loss=0.0150
|
| 282 |
+
step 2800000 | 26502s | train_loss=0.0154 | fid=140.5276
|
| 283 |
+
step 2810000 | 26588s | train_loss=0.0155
|
| 284 |
+
step 2820000 | 26667s | train_loss=0.0163
|
| 285 |
+
step 2830000 | 26754s | train_loss=0.0160
|
| 286 |
+
step 2840000 | 26840s | train_loss=0.0154
|
| 287 |
+
step 2850000 | 26925s | train_loss=0.0156
|
| 288 |
+
step 2860000 | 27011s | train_loss=0.0163
|
| 289 |
+
step 2870000 | 27098s | train_loss=0.0149
|
| 290 |
+
step 2880000 | 27185s | train_loss=0.0154
|
| 291 |
+
step 2890000 | 27266s | train_loss=0.0149
|
| 292 |
+
step 2900000 | 27465s | train_loss=0.0158 | fid=140.8333
|
| 293 |
+
step 2910000 | 27552s | train_loss=0.0147
|
| 294 |
+
step 2920000 | 27624s | train_loss=0.0155
|
| 295 |
+
step 2930000 | 27708s | train_loss=0.0152
|
| 296 |
+
step 2940000 | 27795s | train_loss=0.0149
|
| 297 |
+
step 2950000 | 27879s | train_loss=0.0151
|
| 298 |
+
step 2960000 | 27966s | train_loss=0.0163
|
| 299 |
+
step 2970000 | 28053s | train_loss=0.0157
|
| 300 |
+
step 2980000 | 28140s | train_loss=0.0148
|
| 301 |
+
step 2990000 | 28227s | train_loss=0.0156
|
| 302 |
+
step 3000000 | 28398s | train_loss=0.0152 | fid=141.1816
|
| 303 |
+
step 3010000 | 28485s | train_loss=0.0147
|
| 304 |
+
step 3020000 | 28568s | train_loss=0.0150
|
| 305 |
+
step 3030000 | 28644s | train_loss=0.0151
|
| 306 |
+
step 3040000 | 28731s | train_loss=0.0146
|
| 307 |
+
step 3050000 | 28815s | train_loss=0.0152
|
| 308 |
+
step 3060000 | 28901s | train_loss=0.0155
|
| 309 |
+
step 3070000 | 28988s | train_loss=0.0144
|
| 310 |
+
step 3080000 | 29075s | train_loss=0.0144
|
| 311 |
+
step 3090000 | 29162s | train_loss=0.0142
|
| 312 |
+
step 3100000 | 29351s | train_loss=0.0139 | fid=140.9524
|
| 313 |
+
step 3110000 | 29438s | train_loss=0.0148
|
| 314 |
+
step 3120000 | 29525s | train_loss=0.0141
|
| 315 |
+
step 3130000 | 29607s | train_loss=0.0147
|
| 316 |
+
step 3140000 | 29694s | train_loss=0.0149
|
| 317 |
+
step 3150000 | 29779s | train_loss=0.0141
|
| 318 |
+
step 3160000 | 29865s | train_loss=0.0141
|
| 319 |
+
step 3170000 | 29952s | train_loss=0.0152
|
| 320 |
+
step 3180000 | 30038s | train_loss=0.0140
|
| 321 |
+
step 3190000 | 30125s | train_loss=0.0142
|
| 322 |
+
step 3200000 | 30317s | train_loss=0.0141 | fid=141.1137
|
| 323 |
+
step 3210000 | 30393s | train_loss=0.0141
|
| 324 |
+
step 3220000 | 30474s | train_loss=0.0139
|
| 325 |
+
step 3230000 | 30552s | train_loss=0.0148
|
| 326 |
+
step 3240000 | 30639s | train_loss=0.0145
|
| 327 |
+
step 3250000 | 30726s | train_loss=0.0138
|
| 328 |
+
step 3260000 | 30811s | train_loss=0.0140
|
| 329 |
+
step 3270000 | 30896s | train_loss=0.0143
|
| 330 |
+
step 3280000 | 30983s | train_loss=0.0142
|
| 331 |
+
step 3290000 | 31070s | train_loss=0.0136
|
| 332 |
+
step 3300000 | 31257s | train_loss=0.0138 | fid=140.2438
|
| 333 |
+
step 3310000 | 31341s | train_loss=0.0137
|
| 334 |
+
step 3320000 | 31419s | train_loss=0.0140
|
| 335 |
+
step 3330000 | 31494s | train_loss=0.0140
|
| 336 |
+
step 3340000 | 31581s | train_loss=0.0133
|
| 337 |
+
step 3350000 | 31668s | train_loss=0.0144
|
| 338 |
+
step 3360000 | 31753s | train_loss=0.0140
|
| 339 |
+
step 3370000 | 31840s | train_loss=0.0144
|
| 340 |
+
step 3380000 | 31925s | train_loss=0.0136
|
| 341 |
+
step 3390000 | 32012s | train_loss=0.0132
|
| 342 |
+
step 3400000 | 32212s | train_loss=0.0133 | fid=141.0704
|
| 343 |
+
step 3410000 | 32299s | train_loss=0.0133
|
| 344 |
+
step 3420000 | 32378s | train_loss=0.0133
|
| 345 |
+
step 3430000 | 32448s | train_loss=0.0138
|
| 346 |
+
step 3440000 | 32535s | train_loss=0.0151
|
| 347 |
+
step 3450000 | 32622s | train_loss=0.0132
|
| 348 |
+
step 3460000 | 32708s | train_loss=0.0134
|
| 349 |
+
step 3470000 | 32794s | train_loss=0.0130
|
| 350 |
+
step 3480000 | 32880s | train_loss=0.0132
|
| 351 |
+
step 3490000 | 32967s | train_loss=0.0139
|
| 352 |
+
step 3500000 | 33138s | train_loss=0.0136 | fid=140.9446
|
| 353 |
+
step 3510000 | 33225s | train_loss=0.0133
|
| 354 |
+
step 3520000 | 33312s | train_loss=0.0132
|
| 355 |
+
step 3530000 | 33388s | train_loss=0.0131
|
| 356 |
+
step 3540000 | 33465s | train_loss=0.0133
|
| 357 |
+
step 3550000 | 33552s | train_loss=0.0133
|
| 358 |
+
step 3560000 | 33639s | train_loss=0.0128
|
| 359 |
+
step 3570000 | 33724s | train_loss=0.0136
|
| 360 |
+
step 3580000 | 33811s | train_loss=0.0134
|
| 361 |
+
step 3590000 | 33896s | train_loss=0.0132
|
| 362 |
+
step 3600000 | 34068s | train_loss=0.0131 | fid=141.2266
|
| 363 |
+
step 3610000 | 34155s | train_loss=0.0130
|
| 364 |
+
step 3620000 | 34242s | train_loss=0.0127
|
| 365 |
+
step 3630000 | 34329s | train_loss=0.0130
|
| 366 |
+
step 3640000 | 34402s | train_loss=0.0138
|
| 367 |
+
step 3650000 | 34481s | train_loss=0.0132
|
| 368 |
+
step 3660000 | 34568s | train_loss=0.0131
|
| 369 |
+
step 3670000 | 34656s | train_loss=0.0126
|
| 370 |
+
step 3680000 | 34741s | train_loss=0.0128
|
| 371 |
+
step 3690000 | 34827s | train_loss=0.0122
|
| 372 |
+
step 3700000 | 35028s | train_loss=0.0121 | fid=141.7020
|
| 373 |
+
step 3710000 | 35115s | train_loss=0.0131
|
| 374 |
+
step 3720000 | 35202s | train_loss=0.0128
|
| 375 |
+
step 3730000 | 35288s | train_loss=0.0127
|
| 376 |
+
step 3740000 | 35359s | train_loss=0.0124
|
| 377 |
+
step 3750000 | 35436s | train_loss=0.0131
|
| 378 |
+
step 3760000 | 35518s | train_loss=0.0126
|
| 379 |
+
step 3770000 | 35605s | train_loss=0.0124
|
| 380 |
+
step 3780000 | 35690s | train_loss=0.0125
|
| 381 |
+
step 3790000 | 35777s | train_loss=0.0123
|
| 382 |
+
step 3800000 | 35985s | train_loss=0.0124 | fid=140.8242
|
| 383 |
+
step 3810000 | 36072s | train_loss=0.0118
|
| 384 |
+
step 3820000 | 36159s | train_loss=0.0119
|
| 385 |
+
step 3830000 | 36246s | train_loss=0.0126
|
| 386 |
+
step 3840000 | 36323s | train_loss=0.0119
|
| 387 |
+
step 3850000 | 36410s | train_loss=0.0129
|
| 388 |
+
step 3860000 | 36487s | train_loss=0.0118
|
| 389 |
+
step 3870000 | 36574s | train_loss=0.0122
|
| 390 |
+
step 3880000 | 36660s | train_loss=0.0119
|
| 391 |
+
step 3890000 | 36746s | train_loss=0.0122
|
| 392 |
+
step 3900000 | 36936s | train_loss=0.0123 | fid=140.6319
|
| 393 |
+
step 3910000 | 37023s | train_loss=0.0121
|
| 394 |
+
step 3920000 | 37110s | train_loss=0.0115
|
| 395 |
+
step 3930000 | 37197s | train_loss=0.0121
|
| 396 |
+
step 3940000 | 37272s | train_loss=0.0124
|
| 397 |
+
step 3950000 | 37359s | train_loss=0.0117
|
| 398 |
+
step 3960000 | 37446s | train_loss=0.0121
|
| 399 |
+
step 3970000 | 37520s | train_loss=0.0118
|
| 400 |
+
step 3980000 | 37605s | train_loss=0.0121
|
| 401 |
+
step 3990000 | 37692s | train_loss=0.0119
|
| 402 |
+
step 4000000 | 37899s | train_loss=0.0121 | fid=141.0354
|
| 403 |
+
step 4010000 | 37984s | train_loss=0.0119
|
| 404 |
+
step 4020000 | 38071s | train_loss=0.0125
|
| 405 |
+
step 4030000 | 38158s | train_loss=0.0114
|
| 406 |
+
step 4040000 | 38240s | train_loss=0.0114
|
| 407 |
+
step 4050000 | 38327s | train_loss=0.0120
|
| 408 |
+
step 4060000 | 38414s | train_loss=0.0121
|
| 409 |
+
step 4070000 | 38499s | train_loss=0.0116
|
| 410 |
+
step 4080000 | 38584s | train_loss=0.0121
|
| 411 |
+
step 4090000 | 38671s | train_loss=0.0121
|
| 412 |
+
step 4100000 | 38831s | train_loss=0.0118 | fid=140.7710
|
| 413 |
+
step 4110000 | 38918s | train_loss=0.0116
|
| 414 |
+
step 4120000 | 39004s | train_loss=0.0116
|
| 415 |
+
step 4130000 | 39090s | train_loss=0.0121
|
| 416 |
+
step 4140000 | 39174s | train_loss=0.0116
|
| 417 |
+
step 4150000 | 39261s | train_loss=0.0118
|
| 418 |
+
step 4160000 | 39348s | train_loss=0.0116
|
| 419 |
+
step 4170000 | 39435s | train_loss=0.0119
|
| 420 |
+
step 4180000 | 39519s | train_loss=0.0112
|
| 421 |
+
step 4190000 | 39605s | train_loss=0.0111
|
| 422 |
+
step 4200000 | 39762s | train_loss=0.0116 | fid=140.9288
|
| 423 |
+
step 4210000 | 39849s | train_loss=0.0116
|
| 424 |
+
step 4220000 | 39936s | train_loss=0.0114
|
| 425 |
+
step 4230000 | 40021s | train_loss=0.0115
|
| 426 |
+
step 4240000 | 40105s | train_loss=0.0111
|
| 427 |
+
step 4250000 | 40192s | train_loss=0.0115
|
| 428 |
+
step 4260000 | 40279s | train_loss=0.0114
|
| 429 |
+
step 4270000 | 40366s | train_loss=0.0116
|
| 430 |
+
step 4280000 | 40453s | train_loss=0.0113
|
| 431 |
+
step 4290000 | 40532s | train_loss=0.0115
|
| 432 |
+
step 4300000 | 40749s | train_loss=0.0114 | fid=140.9503
|
| 433 |
+
step 4310000 | 40836s | train_loss=0.0112
|
| 434 |
+
step 4320000 | 40923s | train_loss=0.0109
|
| 435 |
+
step 4330000 | 41006s | train_loss=0.0109
|
| 436 |
+
step 4340000 | 41079s | train_loss=0.0106
|
| 437 |
+
step 4350000 | 41165s | train_loss=0.0108
|
| 438 |
+
step 4360000 | 41252s | train_loss=0.0112
|
| 439 |
+
step 4370000 | 41339s | train_loss=0.0106
|
| 440 |
+
step 4380000 | 41426s | train_loss=0.0111
|
| 441 |
+
step 4390000 | 41506s | train_loss=0.0113
|
| 442 |
+
step 4400000 | 41684s | train_loss=0.0106 | fid=140.9309
|
| 443 |
+
step 4410000 | 41771s | train_loss=0.0112
|
| 444 |
+
step 4420000 | 41858s | train_loss=0.0112
|
| 445 |
+
step 4430000 | 41945s | train_loss=0.0106
|
| 446 |
+
step 4440000 | 42021s | train_loss=0.0108
|
| 447 |
+
step 4450000 | 42103s | train_loss=0.0108
|
| 448 |
+
step 4460000 | 42190s | train_loss=0.0115
|
| 449 |
+
step 4470000 | 42277s | train_loss=0.0114
|
| 450 |
+
step 4480000 | 42364s | train_loss=0.0105
|
| 451 |
+
step 4490000 | 42451s | train_loss=0.0113
|
| 452 |
+
step 4500000 | 42650s | train_loss=0.0104 | fid=141.6680
|
| 453 |
+
step 4510000 | 42736s | train_loss=0.0109
|
| 454 |
+
step 4520000 | 42823s | train_loss=0.0106
|
| 455 |
+
step 4530000 | 42910s | train_loss=0.0107
|
| 456 |
+
step 4540000 | 42985s | train_loss=0.0106
|
| 457 |
+
step 4550000 | 43067s | train_loss=0.0106
|
| 458 |
+
step 4560000 | 43154s | train_loss=0.0107
|
| 459 |
+
step 4570000 | 43241s | train_loss=0.0103
|
| 460 |
+
step 4580000 | 43328s | train_loss=0.0104
|
| 461 |
+
step 4590000 | 43415s | train_loss=0.0105
|
| 462 |
+
step 4600000 | 43604s | train_loss=0.0103 | fid=141.4305
|
| 463 |
+
step 4610000 | 43690s | train_loss=0.0101
|
| 464 |
+
step 4620000 | 43777s | train_loss=0.0107
|
| 465 |
+
step 4630000 | 43864s | train_loss=0.0104
|
| 466 |
+
step 4640000 | 43943s | train_loss=0.0110
|
| 467 |
+
step 4650000 | 44028s | train_loss=0.0112
|
| 468 |
+
step 4660000 | 44113s | train_loss=0.0105
|
| 469 |
+
step 4670000 | 44200s | train_loss=0.0106
|
| 470 |
+
step 4680000 | 44287s | train_loss=0.0098
|
| 471 |
+
step 4690000 | 44374s | train_loss=0.0108
|
| 472 |
+
step 4700000 | 44558s | train_loss=0.0099 | fid=140.9887
|
| 473 |
+
step 4710000 | 44643s | train_loss=0.0101
|
| 474 |
+
step 4720000 | 44729s | train_loss=0.0105
|
| 475 |
+
step 4730000 | 44816s | train_loss=0.0104
|
| 476 |
+
step 4740000 | 44895s | train_loss=0.0105
|
| 477 |
+
step 4750000 | 44976s | train_loss=0.0102
|
| 478 |
+
step 4760000 | 45062s | train_loss=0.0106
|
| 479 |
+
step 4770000 | 45148s | train_loss=0.0107
|
| 480 |
+
step 4780000 | 45235s | train_loss=0.0100
|
| 481 |
+
step 4790000 | 45322s | train_loss=0.0107
|
| 482 |
+
step 4800000 | 45524s | train_loss=0.0101 | fid=141.6229
|
| 483 |
+
step 4810000 | 45596s | train_loss=0.0103
|
| 484 |
+
step 4820000 | 45680s | train_loss=0.0101
|
| 485 |
+
step 4830000 | 45766s | train_loss=0.0104
|
| 486 |
+
step 4840000 | 45847s | train_loss=0.0100
|
| 487 |
+
step 4850000 | 45927s | train_loss=0.0100
|
| 488 |
+
step 4860000 | 46014s | train_loss=0.0099
|
| 489 |
+
step 4870000 | 46100s | train_loss=0.0101
|
| 490 |
+
step 4880000 | 46186s | train_loss=0.0100
|
| 491 |
+
step 4890000 | 46273s | train_loss=0.0099
|
| 492 |
+
step 4900000 | 46469s | train_loss=0.0099 | fid=141.2791
|
| 493 |
+
step 4910000 | 46556s | train_loss=0.0099
|
| 494 |
+
step 4920000 | 46630s | train_loss=0.0100
|
| 495 |
+
step 4930000 | 46716s | train_loss=0.0096
|
| 496 |
+
step 4940000 | 46799s | train_loss=0.0098
|
| 497 |
+
step 4950000 | 46880s | train_loss=0.0102
|
| 498 |
+
step 4960000 | 46967s | train_loss=0.0099
|
| 499 |
+
step 4970000 | 47054s | train_loss=0.0100
|
| 500 |
+
step 4980000 | 47139s | train_loss=0.0096
|
| 501 |
+
step 4990000 | 47226s | train_loss=0.0100
|
| 502 |
+
step 5000000 | 47414s | train_loss=0.0100 | fid=141.2685
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d60/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d60/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d60",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 44,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d60/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d60/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 73s | train_loss=0.9910 | fid=257.2149
|
| 3 |
+
step 10000 | 166s | train_loss=0.8598
|
| 4 |
+
step 20000 | 259s | train_loss=0.7835
|
| 5 |
+
step 30000 | 352s | train_loss=0.6565
|
| 6 |
+
step 40000 | 445s | train_loss=0.4643
|
| 7 |
+
step 50000 | 539s | train_loss=0.2995
|
| 8 |
+
step 60000 | 632s | train_loss=0.2131
|
| 9 |
+
step 70000 | 725s | train_loss=0.1712
|
| 10 |
+
step 80000 | 818s | train_loss=0.1496
|
| 11 |
+
step 90000 | 912s | train_loss=0.1350
|
| 12 |
+
step 100000 | 1073s | train_loss=0.1250 | fid=149.0896
|
| 13 |
+
step 110000 | 1164s | train_loss=0.1177
|
| 14 |
+
step 120000 | 1257s | train_loss=0.1104
|
| 15 |
+
step 130000 | 1350s | train_loss=0.1055
|
| 16 |
+
step 140000 | 1443s | train_loss=0.1005
|
| 17 |
+
step 150000 | 1537s | train_loss=0.0963
|
| 18 |
+
step 160000 | 1630s | train_loss=0.0943
|
| 19 |
+
step 170000 | 1723s | train_loss=0.0876
|
| 20 |
+
step 180000 | 1817s | train_loss=0.0847
|
| 21 |
+
step 190000 | 1910s | train_loss=0.0822
|
| 22 |
+
step 200000 | 2072s | train_loss=0.0820 | fid=142.8028
|
| 23 |
+
step 210000 | 2162s | train_loss=0.0770
|
| 24 |
+
step 220000 | 2255s | train_loss=0.0772
|
| 25 |
+
step 230000 | 2348s | train_loss=0.0735
|
| 26 |
+
step 240000 | 2441s | train_loss=0.0715
|
| 27 |
+
step 250000 | 2534s | train_loss=0.0704
|
| 28 |
+
step 260000 | 2628s | train_loss=0.0693
|
| 29 |
+
step 270000 | 2721s | train_loss=0.0671
|
| 30 |
+
step 280000 | 2814s | train_loss=0.0650
|
| 31 |
+
step 290000 | 2901s | train_loss=0.0632
|
| 32 |
+
step 300000 | 3070s | train_loss=0.0618 | fid=137.8933
|
| 33 |
+
step 310000 | 3161s | train_loss=0.0608
|
| 34 |
+
step 320000 | 3253s | train_loss=0.0597
|
| 35 |
+
step 330000 | 3346s | train_loss=0.0583
|
| 36 |
+
step 340000 | 3439s | train_loss=0.0567
|
| 37 |
+
step 350000 | 3533s | train_loss=0.0576
|
| 38 |
+
step 360000 | 3626s | train_loss=0.0558
|
| 39 |
+
step 370000 | 3719s | train_loss=0.0548
|
| 40 |
+
step 380000 | 3809s | train_loss=0.0546
|
| 41 |
+
step 390000 | 3899s | train_loss=0.0535
|
| 42 |
+
step 400000 | 4067s | train_loss=0.0528 | fid=132.5876
|
| 43 |
+
step 410000 | 4159s | train_loss=0.0513
|
| 44 |
+
step 420000 | 4251s | train_loss=0.0507
|
| 45 |
+
step 430000 | 4344s | train_loss=0.0510
|
| 46 |
+
step 440000 | 4437s | train_loss=0.0496
|
| 47 |
+
step 450000 | 4530s | train_loss=0.0497
|
| 48 |
+
step 460000 | 4623s | train_loss=0.0492
|
| 49 |
+
step 470000 | 4716s | train_loss=0.0477
|
| 50 |
+
step 480000 | 4804s | train_loss=0.0475
|
| 51 |
+
step 490000 | 4897s | train_loss=0.0481
|
| 52 |
+
step 500000 | 5064s | train_loss=0.0472 | fid=131.3928
|
| 53 |
+
step 510000 | 5156s | train_loss=0.0462
|
| 54 |
+
step 520000 | 5249s | train_loss=0.0454
|
| 55 |
+
step 530000 | 5341s | train_loss=0.0451
|
| 56 |
+
step 540000 | 5434s | train_loss=0.0445
|
| 57 |
+
step 550000 | 5527s | train_loss=0.0446
|
| 58 |
+
step 560000 | 5620s | train_loss=0.0435
|
| 59 |
+
step 570000 | 5710s | train_loss=0.0438
|
| 60 |
+
step 580000 | 5800s | train_loss=0.0428
|
| 61 |
+
step 590000 | 5894s | train_loss=0.0424
|
| 62 |
+
step 600000 | 6060s | train_loss=0.0428 | fid=127.7879
|
| 63 |
+
step 610000 | 6152s | train_loss=0.0427
|
| 64 |
+
step 620000 | 6245s | train_loss=0.0425
|
| 65 |
+
step 630000 | 6337s | train_loss=0.0409
|
| 66 |
+
step 640000 | 6430s | train_loss=0.0409
|
| 67 |
+
step 650000 | 6524s | train_loss=0.0416
|
| 68 |
+
step 660000 | 6614s | train_loss=0.0411
|
| 69 |
+
step 670000 | 6704s | train_loss=0.0394
|
| 70 |
+
step 680000 | 6797s | train_loss=0.0400
|
| 71 |
+
step 690000 | 6890s | train_loss=0.0400
|
| 72 |
+
step 700000 | 7058s | train_loss=0.0378 | fid=126.9506
|
| 73 |
+
step 710000 | 7149s | train_loss=0.0385
|
| 74 |
+
step 720000 | 7243s | train_loss=0.0400
|
| 75 |
+
step 730000 | 7336s | train_loss=0.0382
|
| 76 |
+
step 740000 | 7428s | train_loss=0.0377
|
| 77 |
+
step 750000 | 7520s | train_loss=0.0371
|
| 78 |
+
step 760000 | 7609s | train_loss=0.0374
|
| 79 |
+
step 770000 | 7701s | train_loss=0.0372
|
| 80 |
+
step 780000 | 7794s | train_loss=0.0367
|
| 81 |
+
step 790000 | 7887s | train_loss=0.0364
|
| 82 |
+
step 800000 | 8055s | train_loss=0.0376 | fid=125.0831
|
| 83 |
+
step 810000 | 8146s | train_loss=0.0357
|
| 84 |
+
step 820000 | 8240s | train_loss=0.0363
|
| 85 |
+
step 830000 | 8333s | train_loss=0.0376
|
| 86 |
+
step 840000 | 8425s | train_loss=0.0361
|
| 87 |
+
step 850000 | 8515s | train_loss=0.0358
|
| 88 |
+
step 860000 | 8606s | train_loss=0.0361
|
| 89 |
+
step 870000 | 8699s | train_loss=0.0354
|
| 90 |
+
step 880000 | 8792s | train_loss=0.0346
|
| 91 |
+
step 890000 | 8885s | train_loss=0.0360
|
| 92 |
+
step 900000 | 9053s | train_loss=0.0350 | fid=124.6954
|
| 93 |
+
step 910000 | 9144s | train_loss=0.0346
|
| 94 |
+
step 920000 | 9237s | train_loss=0.0351
|
| 95 |
+
step 930000 | 9330s | train_loss=0.0345
|
| 96 |
+
step 940000 | 9421s | train_loss=0.0339
|
| 97 |
+
step 950000 | 9509s | train_loss=0.0337
|
| 98 |
+
step 960000 | 9602s | train_loss=0.0342
|
| 99 |
+
step 970000 | 9695s | train_loss=0.0348
|
| 100 |
+
step 980000 | 9789s | train_loss=0.0334
|
| 101 |
+
step 990000 | 9882s | train_loss=0.0324
|
| 102 |
+
step 1000000 | 10055s | train_loss=0.0337 | fid=122.4091
|
| 103 |
+
step 1010000 | 10147s | train_loss=0.0322
|
| 104 |
+
step 1020000 | 10240s | train_loss=0.0317
|
| 105 |
+
step 1030000 | 10333s | train_loss=0.0322
|
| 106 |
+
step 1040000 | 10424s | train_loss=0.0330
|
| 107 |
+
step 1050000 | 10513s | train_loss=0.0318
|
| 108 |
+
step 1060000 | 10606s | train_loss=0.0317
|
| 109 |
+
step 1070000 | 10699s | train_loss=0.0324
|
| 110 |
+
step 1080000 | 10792s | train_loss=0.0314
|
| 111 |
+
step 1090000 | 10885s | train_loss=0.0310
|
| 112 |
+
step 1100000 | 11057s | train_loss=0.0304 | fid=121.4507
|
| 113 |
+
step 1110000 | 11149s | train_loss=0.0311
|
| 114 |
+
step 1120000 | 11242s | train_loss=0.0302
|
| 115 |
+
step 1130000 | 11332s | train_loss=0.0296
|
| 116 |
+
step 1140000 | 11415s | train_loss=0.0313
|
| 117 |
+
step 1150000 | 11505s | train_loss=0.0305
|
| 118 |
+
step 1160000 | 11598s | train_loss=0.0301
|
| 119 |
+
step 1170000 | 11692s | train_loss=0.0303
|
| 120 |
+
step 1180000 | 11784s | train_loss=0.0301
|
| 121 |
+
step 1190000 | 11878s | train_loss=0.0296
|
| 122 |
+
step 1200000 | 12051s | train_loss=0.0312 | fid=120.9844
|
| 123 |
+
step 1210000 | 12142s | train_loss=0.0297
|
| 124 |
+
step 1220000 | 12233s | train_loss=0.0298
|
| 125 |
+
step 1230000 | 12326s | train_loss=0.0300
|
| 126 |
+
step 1240000 | 12416s | train_loss=0.0294
|
| 127 |
+
step 1250000 | 12510s | train_loss=0.0280
|
| 128 |
+
step 1260000 | 12601s | train_loss=0.0297
|
| 129 |
+
step 1270000 | 12694s | train_loss=0.0293
|
| 130 |
+
step 1280000 | 12787s | train_loss=0.0293
|
| 131 |
+
step 1290000 | 12880s | train_loss=0.0292
|
| 132 |
+
step 1300000 | 13052s | train_loss=0.0283 | fid=119.9280
|
| 133 |
+
step 1310000 | 13140s | train_loss=0.0289
|
| 134 |
+
step 1320000 | 13221s | train_loss=0.0283
|
| 135 |
+
step 1330000 | 13308s | train_loss=0.0291
|
| 136 |
+
step 1340000 | 13395s | train_loss=0.0278
|
| 137 |
+
step 1350000 | 13488s | train_loss=0.0279
|
| 138 |
+
step 1360000 | 13579s | train_loss=0.0273
|
| 139 |
+
step 1370000 | 13672s | train_loss=0.0280
|
| 140 |
+
step 1380000 | 13765s | train_loss=0.0274
|
| 141 |
+
step 1390000 | 13859s | train_loss=0.0273
|
| 142 |
+
step 1400000 | 14061s | train_loss=0.0272 | fid=121.0784
|
| 143 |
+
step 1410000 | 14144s | train_loss=0.0277
|
| 144 |
+
step 1420000 | 14238s | train_loss=0.0268
|
| 145 |
+
step 1430000 | 14329s | train_loss=0.0265
|
| 146 |
+
step 1440000 | 14422s | train_loss=0.0269
|
| 147 |
+
step 1450000 | 14515s | train_loss=0.0264
|
| 148 |
+
step 1460000 | 14606s | train_loss=0.0261
|
| 149 |
+
step 1470000 | 14700s | train_loss=0.0263
|
| 150 |
+
step 1480000 | 14793s | train_loss=0.0263
|
| 151 |
+
step 1490000 | 14886s | train_loss=0.0262
|
| 152 |
+
step 1500000 | 15060s | train_loss=0.0263 | fid=119.8117
|
| 153 |
+
step 1510000 | 15150s | train_loss=0.0258
|
| 154 |
+
step 1520000 | 15241s | train_loss=0.0258
|
| 155 |
+
step 1530000 | 15334s | train_loss=0.0249
|
| 156 |
+
step 1540000 | 15427s | train_loss=0.0261
|
| 157 |
+
step 1550000 | 15520s | train_loss=0.0260
|
| 158 |
+
step 1560000 | 15614s | train_loss=0.0261
|
| 159 |
+
step 1570000 | 15705s | train_loss=0.0250
|
| 160 |
+
step 1580000 | 15799s | train_loss=0.0255
|
| 161 |
+
step 1590000 | 15892s | train_loss=0.0260
|
| 162 |
+
step 1600000 | 16062s | train_loss=0.0249 | fid=118.3621
|
| 163 |
+
step 1610000 | 16149s | train_loss=0.0246
|
| 164 |
+
step 1620000 | 16238s | train_loss=0.0241
|
| 165 |
+
step 1630000 | 16331s | train_loss=0.0246
|
| 166 |
+
step 1640000 | 16424s | train_loss=0.0242
|
| 167 |
+
step 1650000 | 16517s | train_loss=0.0246
|
| 168 |
+
step 1660000 | 16610s | train_loss=0.0250
|
| 169 |
+
step 1670000 | 16702s | train_loss=0.0239
|
| 170 |
+
step 1680000 | 16795s | train_loss=0.0241
|
| 171 |
+
step 1690000 | 16888s | train_loss=0.0243
|
| 172 |
+
step 1700000 | 17061s | train_loss=0.0240 | fid=119.0539
|
| 173 |
+
step 1710000 | 17150s | train_loss=0.0238
|
| 174 |
+
step 1720000 | 17243s | train_loss=0.0238
|
| 175 |
+
step 1730000 | 17336s | train_loss=0.0233
|
| 176 |
+
step 1740000 | 17430s | train_loss=0.0229
|
| 177 |
+
step 1750000 | 17523s | train_loss=0.0233
|
| 178 |
+
step 1760000 | 17616s | train_loss=0.0234
|
| 179 |
+
step 1770000 | 17708s | train_loss=0.0233
|
| 180 |
+
step 1780000 | 17802s | train_loss=0.0230
|
| 181 |
+
step 1790000 | 17893s | train_loss=0.0228
|
| 182 |
+
step 1800000 | 18061s | train_loss=0.0226 | fid=118.4953
|
| 183 |
+
step 1810000 | 18152s | train_loss=0.0226
|
| 184 |
+
step 1820000 | 18245s | train_loss=0.0221
|
| 185 |
+
step 1830000 | 18338s | train_loss=0.0236
|
| 186 |
+
step 1840000 | 18431s | train_loss=0.0227
|
| 187 |
+
step 1850000 | 18524s | train_loss=0.0227
|
| 188 |
+
step 1860000 | 18617s | train_loss=0.0215
|
| 189 |
+
step 1870000 | 18711s | train_loss=0.0216
|
| 190 |
+
step 1880000 | 18802s | train_loss=0.0217
|
| 191 |
+
step 1890000 | 18893s | train_loss=0.0227
|
| 192 |
+
step 1900000 | 19098s | train_loss=0.0222 | fid=118.6937
|
| 193 |
+
step 1910000 | 19191s | train_loss=0.0209
|
| 194 |
+
step 1920000 | 19284s | train_loss=0.0219
|
| 195 |
+
step 1930000 | 19377s | train_loss=0.0213
|
| 196 |
+
step 1940000 | 19471s | train_loss=0.0210
|
| 197 |
+
step 1950000 | 19564s | train_loss=0.0207
|
| 198 |
+
step 1960000 | 19657s | train_loss=0.0212
|
| 199 |
+
step 1970000 | 19748s | train_loss=0.0212
|
| 200 |
+
step 1980000 | 19828s | train_loss=0.0213
|
| 201 |
+
step 1990000 | 19914s | train_loss=0.0211
|
| 202 |
+
step 2000000 | 20102s | train_loss=0.0206 | fid=118.0325
|
| 203 |
+
step 2010000 | 20195s | train_loss=0.0205
|
| 204 |
+
step 2020000 | 20288s | train_loss=0.0208
|
| 205 |
+
step 2030000 | 20381s | train_loss=0.0208
|
| 206 |
+
step 2040000 | 20474s | train_loss=0.0201
|
| 207 |
+
step 2050000 | 20568s | train_loss=0.0201
|
| 208 |
+
step 2060000 | 20661s | train_loss=0.0202
|
| 209 |
+
step 2070000 | 20745s | train_loss=0.0197
|
| 210 |
+
step 2080000 | 20829s | train_loss=0.0211
|
| 211 |
+
step 2090000 | 20909s | train_loss=0.0196
|
| 212 |
+
step 2100000 | 21121s | train_loss=0.0193 | fid=118.4303
|
| 213 |
+
step 2110000 | 21214s | train_loss=0.0194
|
| 214 |
+
step 2120000 | 21307s | train_loss=0.0204
|
| 215 |
+
step 2130000 | 21401s | train_loss=0.0199
|
| 216 |
+
step 2140000 | 21494s | train_loss=0.0208
|
| 217 |
+
step 2150000 | 21587s | train_loss=0.0192
|
| 218 |
+
step 2160000 | 21677s | train_loss=0.0194
|
| 219 |
+
step 2170000 | 21757s | train_loss=0.0202
|
| 220 |
+
step 2180000 | 21846s | train_loss=0.0195
|
| 221 |
+
step 2190000 | 21932s | train_loss=0.0193
|
| 222 |
+
step 2200000 | 22143s | train_loss=0.0194 | fid=118.0939
|
| 223 |
+
step 2210000 | 22236s | train_loss=0.0199
|
| 224 |
+
step 2220000 | 22329s | train_loss=0.0189
|
| 225 |
+
step 2230000 | 22422s | train_loss=0.0190
|
| 226 |
+
step 2240000 | 22516s | train_loss=0.0183
|
| 227 |
+
step 2250000 | 22608s | train_loss=0.0191
|
| 228 |
+
step 2260000 | 22689s | train_loss=0.0185
|
| 229 |
+
step 2270000 | 22781s | train_loss=0.0193
|
| 230 |
+
step 2280000 | 22860s | train_loss=0.0185
|
| 231 |
+
step 2290000 | 22953s | train_loss=0.0190
|
| 232 |
+
step 2300000 | 23156s | train_loss=0.0179 | fid=117.0638
|
| 233 |
+
step 2310000 | 23249s | train_loss=0.0180
|
| 234 |
+
step 2320000 | 23342s | train_loss=0.0183
|
| 235 |
+
step 2330000 | 23435s | train_loss=0.0180
|
| 236 |
+
step 2340000 | 23529s | train_loss=0.0182
|
| 237 |
+
step 2350000 | 23615s | train_loss=0.0177
|
| 238 |
+
step 2360000 | 23699s | train_loss=0.0183
|
| 239 |
+
step 2370000 | 23783s | train_loss=0.0178
|
| 240 |
+
step 2380000 | 23873s | train_loss=0.0188
|
| 241 |
+
step 2390000 | 23965s | train_loss=0.0176
|
| 242 |
+
step 2400000 | 24196s | train_loss=0.0184 | fid=118.4414
|
| 243 |
+
step 2410000 | 24289s | train_loss=0.0170
|
| 244 |
+
step 2420000 | 24382s | train_loss=0.0172
|
| 245 |
+
step 2430000 | 24475s | train_loss=0.0174
|
| 246 |
+
step 2440000 | 24565s | train_loss=0.0173
|
| 247 |
+
step 2450000 | 24650s | train_loss=0.0168
|
| 248 |
+
step 2460000 | 24737s | train_loss=0.0181
|
| 249 |
+
step 2470000 | 24823s | train_loss=0.0172
|
| 250 |
+
step 2480000 | 24915s | train_loss=0.0166
|
| 251 |
+
step 2490000 | 25008s | train_loss=0.0173
|
| 252 |
+
step 2500000 | 25226s | train_loss=0.0172 | fid=118.2416
|
| 253 |
+
step 2510000 | 25320s | train_loss=0.0166
|
| 254 |
+
step 2520000 | 25413s | train_loss=0.0167
|
| 255 |
+
step 2530000 | 25504s | train_loss=0.0162
|
| 256 |
+
step 2540000 | 25585s | train_loss=0.0168
|
| 257 |
+
step 2550000 | 25676s | train_loss=0.0164
|
| 258 |
+
step 2560000 | 25761s | train_loss=0.0168
|
| 259 |
+
step 2570000 | 25852s | train_loss=0.0160
|
| 260 |
+
step 2580000 | 25943s | train_loss=0.0165
|
| 261 |
+
step 2590000 | 26037s | train_loss=0.0163
|
| 262 |
+
step 2600000 | 26259s | train_loss=0.0168 | fid=117.1841
|
| 263 |
+
step 2610000 | 26352s | train_loss=0.0166
|
| 264 |
+
step 2620000 | 26446s | train_loss=0.0158
|
| 265 |
+
step 2630000 | 26528s | train_loss=0.0171
|
| 266 |
+
step 2640000 | 26621s | train_loss=0.0166
|
| 267 |
+
step 2650000 | 26707s | train_loss=0.0159
|
| 268 |
+
step 2660000 | 26799s | train_loss=0.0161
|
| 269 |
+
step 2670000 | 26890s | train_loss=0.0159
|
| 270 |
+
step 2680000 | 26983s | train_loss=0.0162
|
| 271 |
+
step 2690000 | 27077s | train_loss=0.0158
|
| 272 |
+
step 2700000 | 27301s | train_loss=0.0154 | fid=118.1404
|
| 273 |
+
step 2710000 | 27394s | train_loss=0.0159
|
| 274 |
+
step 2720000 | 27473s | train_loss=0.0157
|
| 275 |
+
step 2730000 | 27565s | train_loss=0.0158
|
| 276 |
+
step 2740000 | 27642s | train_loss=0.0150
|
| 277 |
+
step 2750000 | 27735s | train_loss=0.0149
|
| 278 |
+
step 2760000 | 27825s | train_loss=0.0150
|
| 279 |
+
step 2770000 | 27918s | train_loss=0.0152
|
| 280 |
+
step 2780000 | 28011s | train_loss=0.0154
|
| 281 |
+
step 2790000 | 28105s | train_loss=0.0147
|
| 282 |
+
step 2800000 | 28313s | train_loss=0.0148 | fid=117.6060
|
| 283 |
+
step 2810000 | 28397s | train_loss=0.0148
|
| 284 |
+
step 2820000 | 28490s | train_loss=0.0153
|
| 285 |
+
step 2830000 | 28577s | train_loss=0.0151
|
| 286 |
+
step 2840000 | 28662s | train_loss=0.0146
|
| 287 |
+
step 2850000 | 28755s | train_loss=0.0154
|
| 288 |
+
step 2860000 | 28845s | train_loss=0.0152
|
| 289 |
+
step 2870000 | 28938s | train_loss=0.0144
|
| 290 |
+
step 2880000 | 29031s | train_loss=0.0144
|
| 291 |
+
step 2890000 | 29124s | train_loss=0.0145
|
| 292 |
+
step 2900000 | 29324s | train_loss=0.0152 | fid=117.5815
|
| 293 |
+
step 2910000 | 29411s | train_loss=0.0140
|
| 294 |
+
step 2920000 | 29504s | train_loss=0.0142
|
| 295 |
+
step 2930000 | 29593s | train_loss=0.0148
|
| 296 |
+
step 2940000 | 29686s | train_loss=0.0147
|
| 297 |
+
step 2950000 | 29778s | train_loss=0.0141
|
| 298 |
+
step 2960000 | 29869s | train_loss=0.0144
|
| 299 |
+
step 2970000 | 29963s | train_loss=0.0145
|
| 300 |
+
step 2980000 | 30056s | train_loss=0.0137
|
| 301 |
+
step 2990000 | 30149s | train_loss=0.0147
|
| 302 |
+
step 3000000 | 30358s | train_loss=0.0144 | fid=118.1708
|
| 303 |
+
step 3010000 | 30451s | train_loss=0.0143
|
| 304 |
+
step 3020000 | 30530s | train_loss=0.0146
|
| 305 |
+
step 3030000 | 30623s | train_loss=0.0144
|
| 306 |
+
step 3040000 | 30716s | train_loss=0.0142
|
| 307 |
+
step 3050000 | 30807s | train_loss=0.0144
|
| 308 |
+
step 3060000 | 30899s | train_loss=0.0145
|
| 309 |
+
step 3070000 | 30992s | train_loss=0.0136
|
| 310 |
+
step 3080000 | 31085s | train_loss=0.0137
|
| 311 |
+
step 3090000 | 31177s | train_loss=0.0134
|
| 312 |
+
step 3100000 | 31374s | train_loss=0.0141 | fid=117.7986
|
| 313 |
+
step 3110000 | 31460s | train_loss=0.0143
|
| 314 |
+
step 3120000 | 31548s | train_loss=0.0154
|
| 315 |
+
step 3130000 | 31641s | train_loss=0.0139
|
| 316 |
+
step 3140000 | 31733s | train_loss=0.0142
|
| 317 |
+
step 3150000 | 31826s | train_loss=0.0135
|
| 318 |
+
step 3160000 | 31918s | train_loss=0.0132
|
| 319 |
+
step 3170000 | 32011s | train_loss=0.0136
|
| 320 |
+
step 3180000 | 32105s | train_loss=0.0133
|
| 321 |
+
step 3190000 | 32191s | train_loss=0.0136
|
| 322 |
+
step 3200000 | 32391s | train_loss=0.0129 | fid=117.4546
|
| 323 |
+
step 3210000 | 32472s | train_loss=0.0134
|
| 324 |
+
step 3220000 | 32565s | train_loss=0.0132
|
| 325 |
+
step 3230000 | 32658s | train_loss=0.0134
|
| 326 |
+
step 3240000 | 32750s | train_loss=0.0137
|
| 327 |
+
step 3250000 | 32843s | train_loss=0.0132
|
| 328 |
+
step 3260000 | 32934s | train_loss=0.0143
|
| 329 |
+
step 3270000 | 33027s | train_loss=0.0141
|
| 330 |
+
step 3280000 | 33119s | train_loss=0.0130
|
| 331 |
+
step 3290000 | 33207s | train_loss=0.0127
|
| 332 |
+
step 3300000 | 33398s | train_loss=0.0140 | fid=118.2740
|
| 333 |
+
step 3310000 | 33487s | train_loss=0.0126
|
| 334 |
+
step 3320000 | 33580s | train_loss=0.0134
|
| 335 |
+
step 3330000 | 33674s | train_loss=0.0134
|
| 336 |
+
step 3340000 | 33766s | train_loss=0.0125
|
| 337 |
+
step 3350000 | 33859s | train_loss=0.0129
|
| 338 |
+
step 3360000 | 33951s | train_loss=0.0126
|
| 339 |
+
step 3370000 | 34044s | train_loss=0.0129
|
| 340 |
+
step 3380000 | 34132s | train_loss=0.0125
|
| 341 |
+
step 3390000 | 34225s | train_loss=0.0126
|
| 342 |
+
step 3400000 | 34430s | train_loss=0.0123 | fid=117.7425
|
| 343 |
+
step 3410000 | 34523s | train_loss=0.0128
|
| 344 |
+
step 3420000 | 34616s | train_loss=0.0125
|
| 345 |
+
step 3430000 | 34708s | train_loss=0.0126
|
| 346 |
+
step 3440000 | 34801s | train_loss=0.0132
|
| 347 |
+
step 3450000 | 34895s | train_loss=0.0128
|
| 348 |
+
step 3460000 | 34984s | train_loss=0.0128
|
| 349 |
+
step 3470000 | 35067s | train_loss=0.0122
|
| 350 |
+
step 3480000 | 35160s | train_loss=0.0125
|
| 351 |
+
step 3490000 | 35253s | train_loss=0.0126
|
| 352 |
+
step 3500000 | 35455s | train_loss=0.0127 | fid=118.1119
|
| 353 |
+
step 3510000 | 35549s | train_loss=0.0124
|
| 354 |
+
step 3520000 | 35641s | train_loss=0.0124
|
| 355 |
+
step 3530000 | 35734s | train_loss=0.0122
|
| 356 |
+
step 3540000 | 35828s | train_loss=0.0126
|
| 357 |
+
step 3550000 | 35919s | train_loss=0.0128
|
| 358 |
+
step 3560000 | 35998s | train_loss=0.0124
|
| 359 |
+
step 3570000 | 36092s | train_loss=0.0118
|
| 360 |
+
step 3580000 | 36185s | train_loss=0.0127
|
| 361 |
+
step 3590000 | 36277s | train_loss=0.0119
|
| 362 |
+
step 3600000 | 36469s | train_loss=0.0121 | fid=118.2021
|
| 363 |
+
step 3610000 | 36562s | train_loss=0.0131
|
| 364 |
+
step 3620000 | 36654s | train_loss=0.0124
|
| 365 |
+
step 3630000 | 36747s | train_loss=0.0122
|
| 366 |
+
step 3640000 | 36840s | train_loss=0.0118
|
| 367 |
+
step 3650000 | 36924s | train_loss=0.0122
|
| 368 |
+
step 3660000 | 37014s | train_loss=0.0123
|
| 369 |
+
step 3670000 | 37107s | train_loss=0.0117
|
| 370 |
+
step 3680000 | 37200s | train_loss=0.0118
|
| 371 |
+
step 3690000 | 37282s | train_loss=0.0114
|
| 372 |
+
step 3700000 | 37496s | train_loss=0.0119 | fid=117.5790
|
| 373 |
+
step 3710000 | 37588s | train_loss=0.0114
|
| 374 |
+
step 3720000 | 37681s | train_loss=0.0115
|
| 375 |
+
step 3730000 | 37774s | train_loss=0.0120
|
| 376 |
+
step 3740000 | 37862s | train_loss=0.0116
|
| 377 |
+
step 3750000 | 37946s | train_loss=0.0119
|
| 378 |
+
step 3760000 | 38039s | train_loss=0.0117
|
| 379 |
+
step 3770000 | 38132s | train_loss=0.0115
|
| 380 |
+
step 3780000 | 38221s | train_loss=0.0115
|
| 381 |
+
step 3790000 | 38315s | train_loss=0.0114
|
| 382 |
+
step 3800000 | 38483s | train_loss=0.0113 | fid=118.0993
|
| 383 |
+
step 3810000 | 38575s | train_loss=0.0114
|
| 384 |
+
step 3820000 | 38668s | train_loss=0.0109
|
| 385 |
+
step 3830000 | 38762s | train_loss=0.0113
|
| 386 |
+
step 3840000 | 38852s | train_loss=0.0107
|
| 387 |
+
step 3850000 | 38946s | train_loss=0.0113
|
| 388 |
+
step 3860000 | 39038s | train_loss=0.0111
|
| 389 |
+
step 3870000 | 39130s | train_loss=0.0107
|
| 390 |
+
step 3880000 | 39221s | train_loss=0.0112
|
| 391 |
+
step 3890000 | 39315s | train_loss=0.0118
|
| 392 |
+
step 3900000 | 39486s | train_loss=0.0119 | fid=117.8359
|
| 393 |
+
step 3910000 | 39578s | train_loss=0.0112
|
| 394 |
+
step 3920000 | 39671s | train_loss=0.0109
|
| 395 |
+
step 3930000 | 39762s | train_loss=0.0119
|
| 396 |
+
step 3940000 | 39855s | train_loss=0.0114
|
| 397 |
+
step 3950000 | 39948s | train_loss=0.0108
|
| 398 |
+
step 3960000 | 40039s | train_loss=0.0109
|
| 399 |
+
step 3970000 | 40130s | train_loss=0.0111
|
| 400 |
+
step 3980000 | 40224s | train_loss=0.0113
|
| 401 |
+
step 3990000 | 40317s | train_loss=0.0107
|
| 402 |
+
step 4000000 | 40510s | train_loss=0.0111 | fid=118.5307
|
| 403 |
+
step 4010000 | 40602s | train_loss=0.0108
|
| 404 |
+
step 4020000 | 40692s | train_loss=0.0115
|
| 405 |
+
step 4030000 | 40772s | train_loss=0.0103
|
| 406 |
+
step 4040000 | 40865s | train_loss=0.0105
|
| 407 |
+
step 4050000 | 40959s | train_loss=0.0107
|
| 408 |
+
step 4060000 | 41038s | train_loss=0.0110
|
| 409 |
+
step 4070000 | 41128s | train_loss=0.0106
|
| 410 |
+
step 4080000 | 41221s | train_loss=0.0110
|
| 411 |
+
step 4090000 | 41314s | train_loss=0.0108
|
| 412 |
+
step 4100000 | 41516s | train_loss=0.0106 | fid=117.4388
|
| 413 |
+
step 4110000 | 41607s | train_loss=0.0108
|
| 414 |
+
step 4120000 | 41691s | train_loss=0.0106
|
| 415 |
+
step 4130000 | 41784s | train_loss=0.0106
|
| 416 |
+
step 4140000 | 41877s | train_loss=0.0113
|
| 417 |
+
step 4150000 | 41971s | train_loss=0.0111
|
| 418 |
+
step 4160000 | 42049s | train_loss=0.0105
|
| 419 |
+
step 4170000 | 42143s | train_loss=0.0111
|
| 420 |
+
step 4180000 | 42236s | train_loss=0.0108
|
| 421 |
+
step 4190000 | 42330s | train_loss=0.0109
|
| 422 |
+
step 4200000 | 42542s | train_loss=0.0105 | fid=118.3742
|
| 423 |
+
step 4210000 | 42626s | train_loss=0.0106
|
| 424 |
+
step 4220000 | 42714s | train_loss=0.0103
|
| 425 |
+
step 4230000 | 42807s | train_loss=0.0110
|
| 426 |
+
step 4240000 | 42900s | train_loss=0.0102
|
| 427 |
+
step 4250000 | 42983s | train_loss=0.0105
|
| 428 |
+
step 4260000 | 43071s | train_loss=0.0104
|
| 429 |
+
step 4270000 | 43164s | train_loss=0.0103
|
| 430 |
+
step 4280000 | 43257s | train_loss=0.0105
|
| 431 |
+
step 4290000 | 43351s | train_loss=0.0109
|
| 432 |
+
step 4300000 | 43558s | train_loss=0.0104 | fid=118.1717
|
| 433 |
+
step 4310000 | 43641s | train_loss=0.0106
|
| 434 |
+
step 4320000 | 43734s | train_loss=0.0102
|
| 435 |
+
step 4330000 | 43827s | train_loss=0.0103
|
| 436 |
+
step 4340000 | 43918s | train_loss=0.0101
|
| 437 |
+
step 4350000 | 44003s | train_loss=0.0100
|
| 438 |
+
step 4360000 | 44095s | train_loss=0.0102
|
| 439 |
+
step 4370000 | 44188s | train_loss=0.0104
|
| 440 |
+
step 4380000 | 44282s | train_loss=0.0101
|
| 441 |
+
step 4390000 | 44375s | train_loss=0.0112
|
| 442 |
+
step 4400000 | 44568s | train_loss=0.0102 | fid=118.3541
|
| 443 |
+
step 4410000 | 44661s | train_loss=0.0107
|
| 444 |
+
step 4420000 | 44754s | train_loss=0.0103
|
| 445 |
+
step 4430000 | 44848s | train_loss=0.0102
|
| 446 |
+
step 4440000 | 44927s | train_loss=0.0106
|
| 447 |
+
step 4450000 | 45020s | train_loss=0.0098
|
| 448 |
+
step 4460000 | 45112s | train_loss=0.0102
|
| 449 |
+
step 4470000 | 45205s | train_loss=0.0102
|
| 450 |
+
step 4480000 | 45298s | train_loss=0.0105
|
| 451 |
+
step 4490000 | 45391s | train_loss=0.0102
|
| 452 |
+
step 4500000 | 45607s | train_loss=0.0102 | fid=118.0784
|
| 453 |
+
step 4510000 | 45700s | train_loss=0.0100
|
| 454 |
+
step 4520000 | 45794s | train_loss=0.0098
|
| 455 |
+
step 4530000 | 45874s | train_loss=0.0099
|
| 456 |
+
step 4540000 | 45968s | train_loss=0.0097
|
| 457 |
+
step 4550000 | 46061s | train_loss=0.0100
|
| 458 |
+
step 4560000 | 46153s | train_loss=0.0096
|
| 459 |
+
step 4570000 | 46247s | train_loss=0.0097
|
| 460 |
+
step 4580000 | 46340s | train_loss=0.0097
|
| 461 |
+
step 4590000 | 46429s | train_loss=0.0100
|
| 462 |
+
step 4600000 | 46635s | train_loss=0.0096 | fid=118.3866
|
| 463 |
+
step 4610000 | 46728s | train_loss=0.0094
|
| 464 |
+
step 4620000 | 46814s | train_loss=0.0102
|
| 465 |
+
step 4630000 | 46905s | train_loss=0.0098
|
| 466 |
+
step 4640000 | 46999s | train_loss=0.0105
|
| 467 |
+
step 4650000 | 47090s | train_loss=0.0101
|
| 468 |
+
step 4660000 | 47183s | train_loss=0.0102
|
| 469 |
+
step 4670000 | 47277s | train_loss=0.0096
|
| 470 |
+
step 4680000 | 47370s | train_loss=0.0093
|
| 471 |
+
step 4690000 | 47444s | train_loss=0.0102
|
| 472 |
+
step 4700000 | 47592s | train_loss=0.0094 | fid=117.8640
|
| 473 |
+
step 4710000 | 47667s | train_loss=0.0098
|
| 474 |
+
step 4720000 | 47727s | train_loss=0.0101
|
| 475 |
+
step 4730000 | 47784s | train_loss=0.0096
|
| 476 |
+
step 4740000 | 47842s | train_loss=0.0096
|
| 477 |
+
step 4750000 | 47899s | train_loss=0.0100
|
| 478 |
+
step 4760000 | 47954s | train_loss=0.0098
|
| 479 |
+
step 4770000 | 48011s | train_loss=0.0100
|
| 480 |
+
step 4780000 | 48069s | train_loss=0.0097
|
| 481 |
+
step 4790000 | 48123s | train_loss=0.0100
|
| 482 |
+
step 4800000 | 48235s | train_loss=0.0092 | fid=117.7737
|
| 483 |
+
step 4810000 | 48293s | train_loss=0.0093
|
| 484 |
+
step 4820000 | 48350s | train_loss=0.0100
|
| 485 |
+
step 4830000 | 48408s | train_loss=0.0101
|
| 486 |
+
step 4840000 | 48466s | train_loss=0.0094
|
| 487 |
+
step 4850000 | 48523s | train_loss=0.0099
|
| 488 |
+
step 4860000 | 48577s | train_loss=0.0096
|
| 489 |
+
step 4870000 | 48634s | train_loss=0.0098
|
| 490 |
+
step 4880000 | 48692s | train_loss=0.0093
|
| 491 |
+
step 4890000 | 48733s | train_loss=0.0092
|
| 492 |
+
step 4900000 | 48809s | train_loss=0.0088 | fid=118.3113
|
| 493 |
+
step 4910000 | 48848s | train_loss=0.0096
|
| 494 |
+
step 4920000 | 48886s | train_loss=0.0092
|
| 495 |
+
step 4930000 | 48925s | train_loss=0.0089
|
| 496 |
+
step 4940000 | 48963s | train_loss=0.0095
|
| 497 |
+
step 4950000 | 49001s | train_loss=0.0095
|
| 498 |
+
step 4960000 | 49036s | train_loss=0.0094
|
| 499 |
+
step 4970000 | 49074s | train_loss=0.0093
|
| 500 |
+
step 4980000 | 49112s | train_loss=0.0087
|
| 501 |
+
step 4990000 | 49150s | train_loss=0.0091
|
| 502 |
+
step 5000000 | 49226s | train_loss=0.0092 | fid=118.1132
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d80/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d80/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d80",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 44,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d80/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed44/d80/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 71s | train_loss=0.9946 | fid=273.5855
|
| 3 |
+
step 10000 | 156s | train_loss=0.8471
|
| 4 |
+
step 20000 | 243s | train_loss=0.7461
|
| 5 |
+
step 30000 | 331s | train_loss=0.6101
|
| 6 |
+
step 40000 | 419s | train_loss=0.4386
|
| 7 |
+
step 50000 | 506s | train_loss=0.2956
|
| 8 |
+
step 60000 | 594s | train_loss=0.2146
|
| 9 |
+
step 70000 | 682s | train_loss=0.1732
|
| 10 |
+
step 80000 | 769s | train_loss=0.1494
|
| 11 |
+
step 90000 | 857s | train_loss=0.1356
|
| 12 |
+
step 100000 | 1016s | train_loss=0.1254 | fid=152.3614
|
| 13 |
+
step 110000 | 1101s | train_loss=0.1197
|
| 14 |
+
step 120000 | 1187s | train_loss=0.1124
|
| 15 |
+
step 130000 | 1274s | train_loss=0.1067
|
| 16 |
+
step 140000 | 1362s | train_loss=0.1016
|
| 17 |
+
step 150000 | 1449s | train_loss=0.0981
|
| 18 |
+
step 160000 | 1537s | train_loss=0.0961
|
| 19 |
+
step 170000 | 1625s | train_loss=0.0919
|
| 20 |
+
step 180000 | 1713s | train_loss=0.0879
|
| 21 |
+
step 190000 | 1800s | train_loss=0.0854
|
| 22 |
+
step 200000 | 1957s | train_loss=0.0833 | fid=148.9456
|
| 23 |
+
step 210000 | 2046s | train_loss=0.0815
|
| 24 |
+
step 220000 | 2131s | train_loss=0.0817
|
| 25 |
+
step 230000 | 2217s | train_loss=0.0763
|
| 26 |
+
step 240000 | 2304s | train_loss=0.0750
|
| 27 |
+
step 250000 | 2392s | train_loss=0.0726
|
| 28 |
+
step 260000 | 2479s | train_loss=0.0721
|
| 29 |
+
step 270000 | 2567s | train_loss=0.0706
|
| 30 |
+
step 280000 | 2654s | train_loss=0.0689
|
| 31 |
+
step 290000 | 2742s | train_loss=0.0680
|
| 32 |
+
step 300000 | 2900s | train_loss=0.0655 | fid=143.4219
|
| 33 |
+
step 310000 | 2988s | train_loss=0.0645
|
| 34 |
+
step 320000 | 3075s | train_loss=0.0632
|
| 35 |
+
step 330000 | 3161s | train_loss=0.0627
|
| 36 |
+
step 340000 | 3247s | train_loss=0.0617
|
| 37 |
+
step 350000 | 3334s | train_loss=0.0615
|
| 38 |
+
step 360000 | 3422s | train_loss=0.0587
|
| 39 |
+
step 370000 | 3510s | train_loss=0.0581
|
| 40 |
+
step 380000 | 3597s | train_loss=0.0583
|
| 41 |
+
step 390000 | 3685s | train_loss=0.0566
|
| 42 |
+
step 400000 | 3842s | train_loss=0.0566 | fid=135.7246
|
| 43 |
+
step 410000 | 3930s | train_loss=0.0547
|
| 44 |
+
step 420000 | 4018s | train_loss=0.0544
|
| 45 |
+
step 430000 | 4103s | train_loss=0.0534
|
| 46 |
+
step 440000 | 4191s | train_loss=0.0522
|
| 47 |
+
step 450000 | 4277s | train_loss=0.0523
|
| 48 |
+
step 460000 | 4365s | train_loss=0.0519
|
| 49 |
+
step 470000 | 4453s | train_loss=0.0515
|
| 50 |
+
step 480000 | 4540s | train_loss=0.0504
|
| 51 |
+
step 490000 | 4628s | train_loss=0.0511
|
| 52 |
+
step 500000 | 4786s | train_loss=0.0497 | fid=132.1555
|
| 53 |
+
step 510000 | 4874s | train_loss=0.0492
|
| 54 |
+
step 520000 | 4961s | train_loss=0.0488
|
| 55 |
+
step 530000 | 5050s | train_loss=0.0479
|
| 56 |
+
step 540000 | 5134s | train_loss=0.0485
|
| 57 |
+
step 550000 | 5222s | train_loss=0.0469
|
| 58 |
+
step 560000 | 5308s | train_loss=0.0474
|
| 59 |
+
step 570000 | 5396s | train_loss=0.0468
|
| 60 |
+
step 580000 | 5484s | train_loss=0.0455
|
| 61 |
+
step 590000 | 5571s | train_loss=0.0461
|
| 62 |
+
step 600000 | 5729s | train_loss=0.0455 | fid=129.9777
|
| 63 |
+
step 610000 | 5817s | train_loss=0.0444
|
| 64 |
+
step 620000 | 5905s | train_loss=0.0453
|
| 65 |
+
step 630000 | 5992s | train_loss=0.0443
|
| 66 |
+
step 640000 | 6080s | train_loss=0.0449
|
| 67 |
+
step 650000 | 6165s | train_loss=0.0429
|
| 68 |
+
step 660000 | 6253s | train_loss=0.0436
|
| 69 |
+
step 670000 | 6339s | train_loss=0.0418
|
| 70 |
+
step 680000 | 6427s | train_loss=0.0421
|
| 71 |
+
step 690000 | 6514s | train_loss=0.0412
|
| 72 |
+
step 700000 | 6672s | train_loss=0.0419 | fid=130.7159
|
| 73 |
+
step 710000 | 6760s | train_loss=0.0431
|
| 74 |
+
step 720000 | 6848s | train_loss=0.0418
|
| 75 |
+
step 730000 | 6935s | train_loss=0.0412
|
| 76 |
+
step 740000 | 7023s | train_loss=0.0403
|
| 77 |
+
step 750000 | 7108s | train_loss=0.0420
|
| 78 |
+
step 760000 | 7196s | train_loss=0.0402
|
| 79 |
+
step 770000 | 7283s | train_loss=0.0397
|
| 80 |
+
step 780000 | 7370s | train_loss=0.0392
|
| 81 |
+
step 790000 | 7457s | train_loss=0.0387
|
| 82 |
+
step 800000 | 7615s | train_loss=0.0390 | fid=128.7555
|
| 83 |
+
step 810000 | 7703s | train_loss=0.0378
|
| 84 |
+
step 820000 | 7790s | train_loss=0.0382
|
| 85 |
+
step 830000 | 7878s | train_loss=0.0387
|
| 86 |
+
step 840000 | 7966s | train_loss=0.0385
|
| 87 |
+
step 850000 | 8053s | train_loss=0.0386
|
| 88 |
+
step 860000 | 8138s | train_loss=0.0373
|
| 89 |
+
step 870000 | 8226s | train_loss=0.0371
|
| 90 |
+
step 880000 | 8314s | train_loss=0.0363
|
| 91 |
+
step 890000 | 8400s | train_loss=0.0372
|
| 92 |
+
step 900000 | 8555s | train_loss=0.0367 | fid=126.7087
|
| 93 |
+
step 910000 | 8642s | train_loss=0.0361
|
| 94 |
+
step 920000 | 8730s | train_loss=0.0359
|
| 95 |
+
step 930000 | 8818s | train_loss=0.0361
|
| 96 |
+
step 940000 | 8905s | train_loss=0.0360
|
| 97 |
+
step 950000 | 8993s | train_loss=0.0351
|
| 98 |
+
step 960000 | 9080s | train_loss=0.0349
|
| 99 |
+
step 970000 | 9166s | train_loss=0.0352
|
| 100 |
+
step 980000 | 9253s | train_loss=0.0350
|
| 101 |
+
step 990000 | 9341s | train_loss=0.0344
|
| 102 |
+
step 1000000 | 9497s | train_loss=0.0354 | fid=125.8002
|
| 103 |
+
step 1010000 | 9585s | train_loss=0.0339
|
| 104 |
+
step 1020000 | 9673s | train_loss=0.0343
|
| 105 |
+
step 1030000 | 9760s | train_loss=0.0351
|
| 106 |
+
step 1040000 | 9848s | train_loss=0.0350
|
| 107 |
+
step 1050000 | 9936s | train_loss=0.0327
|
| 108 |
+
step 1060000 | 10024s | train_loss=0.0332
|
| 109 |
+
step 1070000 | 10109s | train_loss=0.0326
|
| 110 |
+
step 1080000 | 10195s | train_loss=0.0326
|
| 111 |
+
step 1090000 | 10283s | train_loss=0.0327
|
| 112 |
+
step 1100000 | 10440s | train_loss=0.0328 | fid=125.8488
|
| 113 |
+
step 1110000 | 10526s | train_loss=0.0325
|
| 114 |
+
step 1120000 | 10614s | train_loss=0.0318
|
| 115 |
+
step 1130000 | 10701s | train_loss=0.0323
|
| 116 |
+
step 1140000 | 10789s | train_loss=0.0321
|
| 117 |
+
step 1150000 | 10877s | train_loss=0.0314
|
| 118 |
+
step 1160000 | 10964s | train_loss=0.0312
|
| 119 |
+
step 1170000 | 11051s | train_loss=0.0319
|
| 120 |
+
step 1180000 | 11136s | train_loss=0.0318
|
| 121 |
+
step 1190000 | 11224s | train_loss=0.0299
|
| 122 |
+
step 1200000 | 11422s | train_loss=0.0314 | fid=124.2860
|
| 123 |
+
step 1210000 | 11508s | train_loss=0.0309
|
| 124 |
+
step 1220000 | 11595s | train_loss=0.0303
|
| 125 |
+
step 1230000 | 11683s | train_loss=0.0307
|
| 126 |
+
step 1240000 | 11771s | train_loss=0.0302
|
| 127 |
+
step 1250000 | 11858s | train_loss=0.0292
|
| 128 |
+
step 1260000 | 11946s | train_loss=0.0305
|
| 129 |
+
step 1270000 | 12034s | train_loss=0.0301
|
| 130 |
+
step 1280000 | 12117s | train_loss=0.0305
|
| 131 |
+
step 1290000 | 12202s | train_loss=0.0299
|
| 132 |
+
step 1300000 | 12362s | train_loss=0.0299 | fid=122.6814
|
| 133 |
+
step 1310000 | 12450s | train_loss=0.0290
|
| 134 |
+
step 1320000 | 12537s | train_loss=0.0292
|
| 135 |
+
step 1330000 | 12623s | train_loss=0.0290
|
| 136 |
+
step 1340000 | 12711s | train_loss=0.0287
|
| 137 |
+
step 1350000 | 12799s | train_loss=0.0291
|
| 138 |
+
step 1360000 | 12886s | train_loss=0.0285
|
| 139 |
+
step 1370000 | 12974s | train_loss=0.0283
|
| 140 |
+
step 1380000 | 13059s | train_loss=0.0284
|
| 141 |
+
step 1390000 | 13141s | train_loss=0.0270
|
| 142 |
+
step 1400000 | 13334s | train_loss=0.0284 | fid=123.4796
|
| 143 |
+
step 1410000 | 13422s | train_loss=0.0283
|
| 144 |
+
step 1420000 | 13509s | train_loss=0.0273
|
| 145 |
+
step 1430000 | 13595s | train_loss=0.0275
|
| 146 |
+
step 1440000 | 13683s | train_loss=0.0272
|
| 147 |
+
step 1450000 | 13771s | train_loss=0.0269
|
| 148 |
+
step 1460000 | 13858s | train_loss=0.0283
|
| 149 |
+
step 1470000 | 13946s | train_loss=0.0271
|
| 150 |
+
step 1480000 | 14031s | train_loss=0.0280
|
| 151 |
+
step 1490000 | 14106s | train_loss=0.0277
|
| 152 |
+
step 1500000 | 14261s | train_loss=0.0269 | fid=122.9180
|
| 153 |
+
step 1510000 | 14348s | train_loss=0.0273
|
| 154 |
+
step 1520000 | 14436s | train_loss=0.0272
|
| 155 |
+
step 1530000 | 14524s | train_loss=0.0267
|
| 156 |
+
step 1540000 | 14610s | train_loss=0.0264
|
| 157 |
+
step 1550000 | 14698s | train_loss=0.0272
|
| 158 |
+
step 1560000 | 14785s | train_loss=0.0270
|
| 159 |
+
step 1570000 | 14873s | train_loss=0.0268
|
| 160 |
+
step 1580000 | 14961s | train_loss=0.0252
|
| 161 |
+
step 1590000 | 15049s | train_loss=0.0257
|
| 162 |
+
step 1600000 | 15198s | train_loss=0.0252 | fid=125.0783
|
| 163 |
+
step 1610000 | 15286s | train_loss=0.0258
|
| 164 |
+
step 1620000 | 15373s | train_loss=0.0257
|
| 165 |
+
step 1630000 | 15461s | train_loss=0.0261
|
| 166 |
+
step 1640000 | 15549s | train_loss=0.0245
|
| 167 |
+
step 1650000 | 15635s | train_loss=0.0252
|
| 168 |
+
step 1660000 | 15722s | train_loss=0.0247
|
| 169 |
+
step 1670000 | 15810s | train_loss=0.0247
|
| 170 |
+
step 1680000 | 15898s | train_loss=0.0244
|
| 171 |
+
step 1690000 | 15985s | train_loss=0.0245
|
| 172 |
+
step 1700000 | 16163s | train_loss=0.0241 | fid=123.9318
|
| 173 |
+
step 1710000 | 16251s | train_loss=0.0249
|
| 174 |
+
step 1720000 | 16338s | train_loss=0.0246
|
| 175 |
+
step 1730000 | 16426s | train_loss=0.0238
|
| 176 |
+
step 1740000 | 16514s | train_loss=0.0240
|
| 177 |
+
step 1750000 | 16601s | train_loss=0.0244
|
| 178 |
+
step 1760000 | 16687s | train_loss=0.0242
|
| 179 |
+
step 1770000 | 16775s | train_loss=0.0237
|
| 180 |
+
step 1780000 | 16863s | train_loss=0.0243
|
| 181 |
+
step 1790000 | 16943s | train_loss=0.0238
|
| 182 |
+
step 1800000 | 17089s | train_loss=0.0240 | fid=120.6778
|
| 183 |
+
step 1810000 | 17177s | train_loss=0.0236
|
| 184 |
+
step 1820000 | 17265s | train_loss=0.0233
|
| 185 |
+
step 1830000 | 17352s | train_loss=0.0238
|
| 186 |
+
step 1840000 | 17440s | train_loss=0.0234
|
| 187 |
+
step 1850000 | 17528s | train_loss=0.0230
|
| 188 |
+
step 1860000 | 17616s | train_loss=0.0235
|
| 189 |
+
step 1870000 | 17702s | train_loss=0.0225
|
| 190 |
+
step 1880000 | 17789s | train_loss=0.0233
|
| 191 |
+
step 1890000 | 17877s | train_loss=0.0236
|
| 192 |
+
step 1900000 | 18036s | train_loss=0.0234 | fid=118.7405
|
| 193 |
+
step 1910000 | 18119s | train_loss=0.0227
|
| 194 |
+
step 1920000 | 18206s | train_loss=0.0241
|
| 195 |
+
step 1930000 | 18294s | train_loss=0.0238
|
| 196 |
+
step 1940000 | 18382s | train_loss=0.0218
|
| 197 |
+
step 1950000 | 18469s | train_loss=0.0221
|
| 198 |
+
step 1960000 | 18557s | train_loss=0.0219
|
| 199 |
+
step 1970000 | 18645s | train_loss=0.0219
|
| 200 |
+
step 1980000 | 18731s | train_loss=0.0231
|
| 201 |
+
step 1990000 | 18819s | train_loss=0.0221
|
| 202 |
+
step 2000000 | 18975s | train_loss=0.0217 | fid=120.5797
|
| 203 |
+
step 2010000 | 19061s | train_loss=0.0218
|
| 204 |
+
step 2020000 | 19136s | train_loss=0.0222
|
| 205 |
+
step 2030000 | 19224s | train_loss=0.0220
|
| 206 |
+
step 2040000 | 19311s | train_loss=0.0216
|
| 207 |
+
step 2050000 | 19399s | train_loss=0.0218
|
| 208 |
+
step 2060000 | 19487s | train_loss=0.0216
|
| 209 |
+
step 2070000 | 19574s | train_loss=0.0213
|
| 210 |
+
step 2080000 | 19662s | train_loss=0.0218
|
| 211 |
+
step 2090000 | 19747s | train_loss=0.0209
|
| 212 |
+
step 2100000 | 19941s | train_loss=0.0211 | fid=118.6871
|
| 213 |
+
step 2110000 | 20029s | train_loss=0.0206
|
| 214 |
+
step 2120000 | 20107s | train_loss=0.0215
|
| 215 |
+
step 2130000 | 20194s | train_loss=0.0215
|
| 216 |
+
step 2140000 | 20282s | train_loss=0.0227
|
| 217 |
+
step 2150000 | 20369s | train_loss=0.0206
|
| 218 |
+
step 2160000 | 20457s | train_loss=0.0208
|
| 219 |
+
step 2170000 | 20545s | train_loss=0.0217
|
| 220 |
+
step 2180000 | 20633s | train_loss=0.0213
|
| 221 |
+
step 2190000 | 20717s | train_loss=0.0208
|
| 222 |
+
step 2200000 | 20923s | train_loss=0.0208 | fid=116.6641
|
| 223 |
+
step 2210000 | 21010s | train_loss=0.0211
|
| 224 |
+
step 2220000 | 21091s | train_loss=0.0203
|
| 225 |
+
step 2230000 | 21171s | train_loss=0.0207
|
| 226 |
+
step 2240000 | 21259s | train_loss=0.0201
|
| 227 |
+
step 2250000 | 21346s | train_loss=0.0201
|
| 228 |
+
step 2260000 | 21434s | train_loss=0.0204
|
| 229 |
+
step 2270000 | 21522s | train_loss=0.0211
|
| 230 |
+
step 2280000 | 21609s | train_loss=0.0200
|
| 231 |
+
step 2290000 | 21691s | train_loss=0.0199
|
| 232 |
+
step 2300000 | 21873s | train_loss=0.0199 | fid=117.3503
|
| 233 |
+
step 2310000 | 21961s | train_loss=0.0199
|
| 234 |
+
step 2320000 | 22047s | train_loss=0.0190
|
| 235 |
+
step 2330000 | 22126s | train_loss=0.0198
|
| 236 |
+
step 2340000 | 22210s | train_loss=0.0204
|
| 237 |
+
step 2350000 | 22297s | train_loss=0.0194
|
| 238 |
+
step 2360000 | 22385s | train_loss=0.0198
|
| 239 |
+
step 2370000 | 22473s | train_loss=0.0194
|
| 240 |
+
step 2380000 | 22560s | train_loss=0.0198
|
| 241 |
+
step 2390000 | 22645s | train_loss=0.0197
|
| 242 |
+
step 2400000 | 22843s | train_loss=0.0191 | fid=116.4873
|
| 243 |
+
step 2410000 | 22931s | train_loss=0.0200
|
| 244 |
+
step 2420000 | 23017s | train_loss=0.0186
|
| 245 |
+
step 2430000 | 23106s | train_loss=0.0189
|
| 246 |
+
step 2440000 | 23182s | train_loss=0.0190
|
| 247 |
+
step 2450000 | 23270s | train_loss=0.0189
|
| 248 |
+
step 2460000 | 23358s | train_loss=0.0204
|
| 249 |
+
step 2470000 | 23445s | train_loss=0.0188
|
| 250 |
+
step 2480000 | 23533s | train_loss=0.0186
|
| 251 |
+
step 2490000 | 23614s | train_loss=0.0192
|
| 252 |
+
step 2500000 | 23800s | train_loss=0.0189 | fid=115.3504
|
| 253 |
+
step 2510000 | 23888s | train_loss=0.0182
|
| 254 |
+
step 2520000 | 23974s | train_loss=0.0192
|
| 255 |
+
step 2530000 | 24061s | train_loss=0.0184
|
| 256 |
+
step 2540000 | 24145s | train_loss=0.0181
|
| 257 |
+
step 2550000 | 24220s | train_loss=0.0183
|
| 258 |
+
step 2560000 | 24307s | train_loss=0.0182
|
| 259 |
+
step 2570000 | 24395s | train_loss=0.0179
|
| 260 |
+
step 2580000 | 24483s | train_loss=0.0184
|
| 261 |
+
step 2590000 | 24566s | train_loss=0.0189
|
| 262 |
+
step 2600000 | 24757s | train_loss=0.0184 | fid=114.0960
|
| 263 |
+
step 2610000 | 24843s | train_loss=0.0179
|
| 264 |
+
step 2620000 | 24929s | train_loss=0.0176
|
| 265 |
+
step 2630000 | 25017s | train_loss=0.0182
|
| 266 |
+
step 2640000 | 25104s | train_loss=0.0179
|
| 267 |
+
step 2650000 | 25188s | train_loss=0.0171
|
| 268 |
+
step 2660000 | 25266s | train_loss=0.0173
|
| 269 |
+
step 2670000 | 25353s | train_loss=0.0175
|
| 270 |
+
step 2680000 | 25441s | train_loss=0.0177
|
| 271 |
+
step 2690000 | 25523s | train_loss=0.0175
|
| 272 |
+
step 2700000 | 25712s | train_loss=0.0172 | fid=113.7057
|
| 273 |
+
step 2710000 | 25798s | train_loss=0.0181
|
| 274 |
+
step 2720000 | 25884s | train_loss=0.0174
|
| 275 |
+
step 2730000 | 25972s | train_loss=0.0172
|
| 276 |
+
step 2740000 | 26059s | train_loss=0.0169
|
| 277 |
+
step 2750000 | 26147s | train_loss=0.0166
|
| 278 |
+
step 2760000 | 26228s | train_loss=0.0173
|
| 279 |
+
step 2770000 | 26307s | train_loss=0.0170
|
| 280 |
+
step 2780000 | 26395s | train_loss=0.0177
|
| 281 |
+
step 2790000 | 26478s | train_loss=0.0164
|
| 282 |
+
step 2800000 | 26655s | train_loss=0.0169 | fid=112.2892
|
| 283 |
+
step 2810000 | 26743s | train_loss=0.0168
|
| 284 |
+
step 2820000 | 26829s | train_loss=0.0177
|
| 285 |
+
step 2830000 | 26915s | train_loss=0.0177
|
| 286 |
+
step 2840000 | 27003s | train_loss=0.0164
|
| 287 |
+
step 2850000 | 27091s | train_loss=0.0168
|
| 288 |
+
step 2860000 | 27178s | train_loss=0.0165
|
| 289 |
+
step 2870000 | 27261s | train_loss=0.0162
|
| 290 |
+
step 2880000 | 27339s | train_loss=0.0162
|
| 291 |
+
step 2890000 | 27422s | train_loss=0.0158
|
| 292 |
+
step 2900000 | 27634s | train_loss=0.0168 | fid=113.2245
|
| 293 |
+
step 2910000 | 27722s | train_loss=0.0155
|
| 294 |
+
step 2920000 | 27808s | train_loss=0.0152
|
| 295 |
+
step 2930000 | 27894s | train_loss=0.0157
|
| 296 |
+
step 2940000 | 27982s | train_loss=0.0161
|
| 297 |
+
step 2950000 | 28070s | train_loss=0.0155
|
| 298 |
+
step 2960000 | 28157s | train_loss=0.0166
|
| 299 |
+
step 2970000 | 28245s | train_loss=0.0164
|
| 300 |
+
step 2980000 | 28321s | train_loss=0.0155
|
| 301 |
+
step 2990000 | 28400s | train_loss=0.0167
|
| 302 |
+
step 3000000 | 28613s | train_loss=0.0162 | fid=111.8433
|
| 303 |
+
step 3010000 | 28701s | train_loss=0.0169
|
| 304 |
+
step 3020000 | 28788s | train_loss=0.0156
|
| 305 |
+
step 3030000 | 28873s | train_loss=0.0152
|
| 306 |
+
step 3040000 | 28961s | train_loss=0.0159
|
| 307 |
+
step 3050000 | 29048s | train_loss=0.0159
|
| 308 |
+
step 3060000 | 29136s | train_loss=0.0158
|
| 309 |
+
step 3070000 | 29224s | train_loss=0.0150
|
| 310 |
+
step 3080000 | 29306s | train_loss=0.0153
|
| 311 |
+
step 3090000 | 29375s | train_loss=0.0149
|
| 312 |
+
step 3100000 | 29542s | train_loss=0.0153 | fid=111.0678
|
| 313 |
+
step 3110000 | 29629s | train_loss=0.0156
|
| 314 |
+
step 3120000 | 29717s | train_loss=0.0160
|
| 315 |
+
step 3130000 | 29803s | train_loss=0.0146
|
| 316 |
+
step 3140000 | 29889s | train_loss=0.0154
|
| 317 |
+
step 3150000 | 29977s | train_loss=0.0154
|
| 318 |
+
step 3160000 | 30065s | train_loss=0.0146
|
| 319 |
+
step 3170000 | 30152s | train_loss=0.0151
|
| 320 |
+
step 3180000 | 30240s | train_loss=0.0148
|
| 321 |
+
step 3190000 | 30314s | train_loss=0.0151
|
| 322 |
+
step 3200000 | 30514s | train_loss=0.0149 | fid=110.6600
|
| 323 |
+
step 3210000 | 30602s | train_loss=0.0147
|
| 324 |
+
step 3220000 | 30689s | train_loss=0.0146
|
| 325 |
+
step 3230000 | 30775s | train_loss=0.0151
|
| 326 |
+
step 3240000 | 30862s | train_loss=0.0146
|
| 327 |
+
step 3250000 | 30949s | train_loss=0.0148
|
| 328 |
+
step 3260000 | 31037s | train_loss=0.0149
|
| 329 |
+
step 3270000 | 31124s | train_loss=0.0148
|
| 330 |
+
step 3280000 | 31212s | train_loss=0.0144
|
| 331 |
+
step 3290000 | 31290s | train_loss=0.0144
|
| 332 |
+
step 3300000 | 31478s | train_loss=0.0142 | fid=110.1298
|
| 333 |
+
step 3310000 | 31565s | train_loss=0.0146
|
| 334 |
+
step 3320000 | 31653s | train_loss=0.0148
|
| 335 |
+
step 3330000 | 31739s | train_loss=0.0147
|
| 336 |
+
step 3340000 | 31827s | train_loss=0.0136
|
| 337 |
+
step 3350000 | 31913s | train_loss=0.0152
|
| 338 |
+
step 3360000 | 32001s | train_loss=0.0141
|
| 339 |
+
step 3370000 | 32089s | train_loss=0.0144
|
| 340 |
+
step 3380000 | 32173s | train_loss=0.0138
|
| 341 |
+
step 3390000 | 32251s | train_loss=0.0137
|
| 342 |
+
step 3400000 | 32446s | train_loss=0.0143 | fid=110.4669
|
| 343 |
+
step 3410000 | 32534s | train_loss=0.0138
|
| 344 |
+
step 3420000 | 32622s | train_loss=0.0139
|
| 345 |
+
step 3430000 | 32708s | train_loss=0.0135
|
| 346 |
+
step 3440000 | 32795s | train_loss=0.0146
|
| 347 |
+
step 3450000 | 32882s | train_loss=0.0130
|
| 348 |
+
step 3460000 | 32969s | train_loss=0.0145
|
| 349 |
+
step 3470000 | 33057s | train_loss=0.0131
|
| 350 |
+
step 3480000 | 33138s | train_loss=0.0140
|
| 351 |
+
step 3490000 | 33226s | train_loss=0.0139
|
| 352 |
+
step 3500000 | 33417s | train_loss=0.0139 | fid=110.1954
|
| 353 |
+
step 3510000 | 33504s | train_loss=0.0142
|
| 354 |
+
step 3520000 | 33592s | train_loss=0.0133
|
| 355 |
+
step 3530000 | 33679s | train_loss=0.0137
|
| 356 |
+
step 3540000 | 33766s | train_loss=0.0150
|
| 357 |
+
step 3550000 | 33854s | train_loss=0.0139
|
| 358 |
+
step 3560000 | 33940s | train_loss=0.0142
|
| 359 |
+
step 3570000 | 34028s | train_loss=0.0135
|
| 360 |
+
step 3580000 | 34109s | train_loss=0.0140
|
| 361 |
+
step 3590000 | 34197s | train_loss=0.0129
|
| 362 |
+
step 3600000 | 34380s | train_loss=0.0133 | fid=110.0562
|
| 363 |
+
step 3610000 | 34455s | train_loss=0.0149
|
| 364 |
+
step 3620000 | 34543s | train_loss=0.0137
|
| 365 |
+
step 3630000 | 34631s | train_loss=0.0136
|
| 366 |
+
step 3640000 | 34717s | train_loss=0.0130
|
| 367 |
+
step 3650000 | 34804s | train_loss=0.0131
|
| 368 |
+
step 3660000 | 34892s | train_loss=0.0128
|
| 369 |
+
step 3670000 | 34977s | train_loss=0.0130
|
| 370 |
+
step 3680000 | 35053s | train_loss=0.0135
|
| 371 |
+
step 3690000 | 35140s | train_loss=0.0128
|
| 372 |
+
step 3700000 | 35361s | train_loss=0.0131 | fid=110.2954
|
| 373 |
+
step 3710000 | 35439s | train_loss=0.0133
|
| 374 |
+
step 3720000 | 35522s | train_loss=0.0133
|
| 375 |
+
step 3730000 | 35610s | train_loss=0.0132
|
| 376 |
+
step 3740000 | 35696s | train_loss=0.0130
|
| 377 |
+
step 3750000 | 35784s | train_loss=0.0136
|
| 378 |
+
step 3760000 | 35872s | train_loss=0.0134
|
| 379 |
+
step 3770000 | 35951s | train_loss=0.0127
|
| 380 |
+
step 3780000 | 36030s | train_loss=0.0129
|
| 381 |
+
step 3790000 | 36118s | train_loss=0.0127
|
| 382 |
+
step 3800000 | 36309s | train_loss=0.0130 | fid=109.3970
|
| 383 |
+
step 3810000 | 36397s | train_loss=0.0127
|
| 384 |
+
step 3820000 | 36475s | train_loss=0.0133
|
| 385 |
+
step 3830000 | 36563s | train_loss=0.0124
|
| 386 |
+
step 3840000 | 36649s | train_loss=0.0123
|
| 387 |
+
step 3850000 | 36736s | train_loss=0.0123
|
| 388 |
+
step 3860000 | 36824s | train_loss=0.0132
|
| 389 |
+
step 3870000 | 36909s | train_loss=0.0131
|
| 390 |
+
step 3880000 | 36987s | train_loss=0.0126
|
| 391 |
+
step 3890000 | 37074s | train_loss=0.0126
|
| 392 |
+
step 3900000 | 37272s | train_loss=0.0127 | fid=108.7941
|
| 393 |
+
step 3910000 | 37359s | train_loss=0.0123
|
| 394 |
+
step 3920000 | 37447s | train_loss=0.0123
|
| 395 |
+
step 3930000 | 37522s | train_loss=0.0126
|
| 396 |
+
step 3940000 | 37608s | train_loss=0.0131
|
| 397 |
+
step 3950000 | 37696s | train_loss=0.0127
|
| 398 |
+
step 3960000 | 37784s | train_loss=0.0127
|
| 399 |
+
step 3970000 | 37865s | train_loss=0.0121
|
| 400 |
+
step 3980000 | 37943s | train_loss=0.0124
|
| 401 |
+
step 3990000 | 38031s | train_loss=0.0124
|
| 402 |
+
step 4000000 | 38198s | train_loss=0.0124 | fid=109.4542
|
| 403 |
+
step 4010000 | 38285s | train_loss=0.0118
|
| 404 |
+
step 4020000 | 38373s | train_loss=0.0127
|
| 405 |
+
step 4030000 | 38461s | train_loss=0.0118
|
| 406 |
+
step 4040000 | 38547s | train_loss=0.0121
|
| 407 |
+
step 4050000 | 38633s | train_loss=0.0124
|
| 408 |
+
step 4060000 | 38721s | train_loss=0.0122
|
| 409 |
+
step 4070000 | 38808s | train_loss=0.0124
|
| 410 |
+
step 4080000 | 38893s | train_loss=0.0127
|
| 411 |
+
step 4090000 | 38980s | train_loss=0.0121
|
| 412 |
+
step 4100000 | 39140s | train_loss=0.0117 | fid=108.2359
|
| 413 |
+
step 4110000 | 39228s | train_loss=0.0116
|
| 414 |
+
step 4120000 | 39315s | train_loss=0.0122
|
| 415 |
+
step 4130000 | 39403s | train_loss=0.0117
|
| 416 |
+
step 4140000 | 39488s | train_loss=0.0127
|
| 417 |
+
step 4150000 | 39576s | train_loss=0.0118
|
| 418 |
+
step 4160000 | 39662s | train_loss=0.0113
|
| 419 |
+
step 4170000 | 39750s | train_loss=0.0120
|
| 420 |
+
step 4180000 | 39835s | train_loss=0.0117
|
| 421 |
+
step 4190000 | 39923s | train_loss=0.0120
|
| 422 |
+
step 4200000 | 40080s | train_loss=0.0115 | fid=108.8862
|
| 423 |
+
step 4210000 | 40167s | train_loss=0.0123
|
| 424 |
+
step 4220000 | 40255s | train_loss=0.0111
|
| 425 |
+
step 4230000 | 40343s | train_loss=0.0119
|
| 426 |
+
step 4240000 | 40431s | train_loss=0.0116
|
| 427 |
+
step 4250000 | 40511s | train_loss=0.0117
|
| 428 |
+
step 4260000 | 40597s | train_loss=0.0117
|
| 429 |
+
step 4270000 | 40683s | train_loss=0.0117
|
| 430 |
+
step 4280000 | 40755s | train_loss=0.0119
|
| 431 |
+
step 4290000 | 40843s | train_loss=0.0123
|
| 432 |
+
step 4300000 | 41055s | train_loss=0.0116 | fid=108.4416
|
| 433 |
+
step 4310000 | 41143s | train_loss=0.0112
|
| 434 |
+
step 4320000 | 41231s | train_loss=0.0112
|
| 435 |
+
step 4330000 | 41319s | train_loss=0.0114
|
| 436 |
+
step 4340000 | 41406s | train_loss=0.0112
|
| 437 |
+
step 4350000 | 41491s | train_loss=0.0111
|
| 438 |
+
step 4360000 | 41571s | train_loss=0.0118
|
| 439 |
+
step 4370000 | 41656s | train_loss=0.0112
|
| 440 |
+
step 4380000 | 41737s | train_loss=0.0112
|
| 441 |
+
step 4390000 | 41825s | train_loss=0.0119
|
| 442 |
+
step 4400000 | 42028s | train_loss=0.0114 | fid=108.6200
|
| 443 |
+
step 4410000 | 42115s | train_loss=0.0122
|
| 444 |
+
step 4420000 | 42203s | train_loss=0.0114
|
| 445 |
+
step 4430000 | 42291s | train_loss=0.0112
|
| 446 |
+
step 4440000 | 42378s | train_loss=0.0112
|
| 447 |
+
step 4450000 | 42466s | train_loss=0.0109
|
| 448 |
+
step 4460000 | 42539s | train_loss=0.0116
|
| 449 |
+
step 4470000 | 42618s | train_loss=0.0112
|
| 450 |
+
step 4480000 | 42698s | train_loss=0.0111
|
| 451 |
+
step 4490000 | 42786s | train_loss=0.0115
|
| 452 |
+
step 4500000 | 42997s | train_loss=0.0110 | fid=107.9440
|
| 453 |
+
step 4510000 | 43083s | train_loss=0.0114
|
| 454 |
+
step 4520000 | 43171s | train_loss=0.0114
|
| 455 |
+
step 4530000 | 43259s | train_loss=0.0109
|
| 456 |
+
step 4540000 | 43346s | train_loss=0.0116
|
| 457 |
+
step 4550000 | 43434s | train_loss=0.0109
|
| 458 |
+
step 4560000 | 43519s | train_loss=0.0118
|
| 459 |
+
step 4570000 | 43586s | train_loss=0.0111
|
| 460 |
+
step 4580000 | 43670s | train_loss=0.0111
|
| 461 |
+
step 4590000 | 43758s | train_loss=0.0114
|
| 462 |
+
step 4600000 | 43949s | train_loss=0.0106 | fid=108.3788
|
| 463 |
+
step 4610000 | 44037s | train_loss=0.0105
|
| 464 |
+
step 4620000 | 44123s | train_loss=0.0117
|
| 465 |
+
step 4630000 | 44210s | train_loss=0.0112
|
| 466 |
+
step 4640000 | 44298s | train_loss=0.0119
|
| 467 |
+
step 4650000 | 44386s | train_loss=0.0120
|
| 468 |
+
step 4660000 | 44474s | train_loss=0.0109
|
| 469 |
+
step 4670000 | 44547s | train_loss=0.0109
|
| 470 |
+
step 4680000 | 44625s | train_loss=0.0106
|
| 471 |
+
step 4690000 | 44712s | train_loss=0.0112
|
| 472 |
+
step 4700000 | 44917s | train_loss=0.0105 | fid=108.1628
|
| 473 |
+
step 4710000 | 45004s | train_loss=0.0108
|
| 474 |
+
step 4720000 | 45091s | train_loss=0.0117
|
| 475 |
+
step 4730000 | 45178s | train_loss=0.0113
|
| 476 |
+
step 4740000 | 45266s | train_loss=0.0111
|
| 477 |
+
step 4750000 | 45354s | train_loss=0.0109
|
| 478 |
+
step 4760000 | 45441s | train_loss=0.0113
|
| 479 |
+
step 4770000 | 45516s | train_loss=0.0114
|
| 480 |
+
step 4780000 | 45589s | train_loss=0.0114
|
| 481 |
+
step 4790000 | 45672s | train_loss=0.0116
|
| 482 |
+
step 4800000 | 45874s | train_loss=0.0106 | fid=108.3481
|
| 483 |
+
step 4810000 | 45962s | train_loss=0.0108
|
| 484 |
+
step 4820000 | 46050s | train_loss=0.0117
|
| 485 |
+
step 4830000 | 46135s | train_loss=0.0113
|
| 486 |
+
step 4840000 | 46223s | train_loss=0.0112
|
| 487 |
+
step 4850000 | 46311s | train_loss=0.0112
|
| 488 |
+
step 4860000 | 46399s | train_loss=0.0104
|
| 489 |
+
step 4870000 | 46473s | train_loss=0.0108
|
| 490 |
+
step 4880000 | 46561s | train_loss=0.0106
|
| 491 |
+
step 4890000 | 46635s | train_loss=0.0105
|
| 492 |
+
step 4900000 | 46821s | train_loss=0.0103 | fid=108.7950
|
| 493 |
+
step 4910000 | 46909s | train_loss=0.0110
|
| 494 |
+
step 4920000 | 46997s | train_loss=0.0105
|
| 495 |
+
step 4930000 | 47083s | train_loss=0.0102
|
| 496 |
+
step 4940000 | 47170s | train_loss=0.0103
|
| 497 |
+
step 4950000 | 47258s | train_loss=0.0103
|
| 498 |
+
step 4960000 | 47346s | train_loss=0.0102
|
| 499 |
+
step 4970000 | 47420s | train_loss=0.0104
|
| 500 |
+
step 4980000 | 47491s | train_loss=0.0101
|
| 501 |
+
step 4990000 | 47562s | train_loss=0.0103
|
| 502 |
+
step 5000000 | 47686s | train_loss=0.0107 | fid=108.1380
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d100/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d100/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d100",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 45,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d100/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d100/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 69s | train_loss=1.0037 | fid=311.3592
|
| 3 |
+
step 10000 | 155s | train_loss=0.8267
|
| 4 |
+
step 20000 | 242s | train_loss=0.7068
|
| 5 |
+
step 30000 | 329s | train_loss=0.5705
|
| 6 |
+
step 40000 | 416s | train_loss=0.4222
|
| 7 |
+
step 50000 | 503s | train_loss=0.2978
|
| 8 |
+
step 60000 | 590s | train_loss=0.2199
|
| 9 |
+
step 70000 | 676s | train_loss=0.1802
|
| 10 |
+
step 80000 | 763s | train_loss=0.1514
|
| 11 |
+
step 90000 | 850s | train_loss=0.1408
|
| 12 |
+
step 100000 | 1001s | train_loss=0.1261 | fid=166.6604
|
| 13 |
+
step 110000 | 1085s | train_loss=0.1181
|
| 14 |
+
step 120000 | 1172s | train_loss=0.1107
|
| 15 |
+
step 130000 | 1258s | train_loss=0.1082
|
| 16 |
+
step 140000 | 1345s | train_loss=0.1024
|
| 17 |
+
step 150000 | 1432s | train_loss=0.0980
|
| 18 |
+
step 160000 | 1519s | train_loss=0.0948
|
| 19 |
+
step 170000 | 1605s | train_loss=0.0911
|
| 20 |
+
step 180000 | 1692s | train_loss=0.0907
|
| 21 |
+
step 190000 | 1779s | train_loss=0.0861
|
| 22 |
+
step 200000 | 1930s | train_loss=0.0859 | fid=159.3363
|
| 23 |
+
step 210000 | 2018s | train_loss=0.0849
|
| 24 |
+
step 220000 | 2101s | train_loss=0.0799
|
| 25 |
+
step 230000 | 2188s | train_loss=0.0774
|
| 26 |
+
step 240000 | 2274s | train_loss=0.0781
|
| 27 |
+
step 250000 | 2361s | train_loss=0.0783
|
| 28 |
+
step 260000 | 2448s | train_loss=0.0733
|
| 29 |
+
step 270000 | 2535s | train_loss=0.0706
|
| 30 |
+
step 280000 | 2621s | train_loss=0.0697
|
| 31 |
+
step 290000 | 2708s | train_loss=0.0693
|
| 32 |
+
step 300000 | 2859s | train_loss=0.0688 | fid=151.6692
|
| 33 |
+
step 310000 | 2946s | train_loss=0.0672
|
| 34 |
+
step 320000 | 3030s | train_loss=0.0661
|
| 35 |
+
step 330000 | 3117s | train_loss=0.0648
|
| 36 |
+
step 340000 | 3203s | train_loss=0.0663
|
| 37 |
+
step 350000 | 3290s | train_loss=0.0616
|
| 38 |
+
step 360000 | 3377s | train_loss=0.0613
|
| 39 |
+
step 370000 | 3464s | train_loss=0.0588
|
| 40 |
+
step 380000 | 3550s | train_loss=0.0592
|
| 41 |
+
step 390000 | 3637s | train_loss=0.0597
|
| 42 |
+
step 400000 | 3788s | train_loss=0.0590 | fid=145.6547
|
| 43 |
+
step 410000 | 3875s | train_loss=0.0570
|
| 44 |
+
step 420000 | 3962s | train_loss=0.0563
|
| 45 |
+
step 430000 | 4046s | train_loss=0.0551
|
| 46 |
+
step 440000 | 4132s | train_loss=0.0563
|
| 47 |
+
step 450000 | 4219s | train_loss=0.0562
|
| 48 |
+
step 460000 | 4306s | train_loss=0.0546
|
| 49 |
+
step 470000 | 4393s | train_loss=0.0546
|
| 50 |
+
step 480000 | 4479s | train_loss=0.0533
|
| 51 |
+
step 490000 | 4566s | train_loss=0.0516
|
| 52 |
+
step 500000 | 4715s | train_loss=0.0523 | fid=142.3393
|
| 53 |
+
step 510000 | 4802s | train_loss=0.0523
|
| 54 |
+
step 520000 | 4889s | train_loss=0.0505
|
| 55 |
+
step 530000 | 4975s | train_loss=0.0515
|
| 56 |
+
step 540000 | 5060s | train_loss=0.0497
|
| 57 |
+
step 550000 | 5146s | train_loss=0.0502
|
| 58 |
+
step 560000 | 5232s | train_loss=0.0491
|
| 59 |
+
step 570000 | 5318s | train_loss=0.0488
|
| 60 |
+
step 580000 | 5404s | train_loss=0.0495
|
| 61 |
+
step 590000 | 5490s | train_loss=0.0480
|
| 62 |
+
step 600000 | 5640s | train_loss=0.0479 | fid=148.3716
|
| 63 |
+
step 610000 | 5727s | train_loss=0.0469
|
| 64 |
+
step 620000 | 5814s | train_loss=0.0487
|
| 65 |
+
step 630000 | 5901s | train_loss=0.0480
|
| 66 |
+
step 640000 | 5985s | train_loss=0.0460
|
| 67 |
+
step 650000 | 6071s | train_loss=0.0457
|
| 68 |
+
step 660000 | 6158s | train_loss=0.0460
|
| 69 |
+
step 670000 | 6245s | train_loss=0.0463
|
| 70 |
+
step 680000 | 6332s | train_loss=0.0461
|
| 71 |
+
step 690000 | 6418s | train_loss=0.0448
|
| 72 |
+
step 700000 | 6570s | train_loss=0.0442 | fid=159.1900
|
| 73 |
+
step 710000 | 6657s | train_loss=0.0438
|
| 74 |
+
step 720000 | 6744s | train_loss=0.0447
|
| 75 |
+
step 730000 | 6830s | train_loss=0.0449
|
| 76 |
+
step 740000 | 6916s | train_loss=0.0429
|
| 77 |
+
step 750000 | 7001s | train_loss=0.0434
|
| 78 |
+
step 760000 | 7088s | train_loss=0.0432
|
| 79 |
+
step 770000 | 7175s | train_loss=0.0428
|
| 80 |
+
step 780000 | 7262s | train_loss=0.0438
|
| 81 |
+
step 790000 | 7348s | train_loss=0.0424
|
| 82 |
+
step 800000 | 7499s | train_loss=0.0424 | fid=180.3393
|
| 83 |
+
step 810000 | 7586s | train_loss=0.0405
|
| 84 |
+
step 820000 | 7673s | train_loss=0.0421
|
| 85 |
+
step 830000 | 7759s | train_loss=0.0412
|
| 86 |
+
step 840000 | 7846s | train_loss=0.0410
|
| 87 |
+
step 850000 | 7930s | train_loss=0.0424
|
| 88 |
+
step 860000 | 8017s | train_loss=0.0392
|
| 89 |
+
step 870000 | 8104s | train_loss=0.0420
|
| 90 |
+
step 880000 | 8190s | train_loss=0.0410
|
| 91 |
+
step 890000 | 8277s | train_loss=0.0398
|
| 92 |
+
step 900000 | 8430s | train_loss=0.0414 | fid=187.2796
|
| 93 |
+
step 910000 | 8517s | train_loss=0.0412
|
| 94 |
+
step 920000 | 8604s | train_loss=0.0395
|
| 95 |
+
step 930000 | 8690s | train_loss=0.0385
|
| 96 |
+
step 940000 | 8777s | train_loss=0.0408
|
| 97 |
+
step 950000 | 8863s | train_loss=0.0389
|
| 98 |
+
step 960000 | 8948s | train_loss=0.0395
|
| 99 |
+
step 970000 | 9035s | train_loss=0.0377
|
| 100 |
+
step 980000 | 9122s | train_loss=0.0378
|
| 101 |
+
step 990000 | 9208s | train_loss=0.0374
|
| 102 |
+
step 1000000 | 9361s | train_loss=0.0382 | fid=187.8301
|
| 103 |
+
step 1010000 | 9448s | train_loss=0.0385
|
| 104 |
+
step 1020000 | 9535s | train_loss=0.0377
|
| 105 |
+
step 1030000 | 9622s | train_loss=0.0382
|
| 106 |
+
step 1040000 | 9708s | train_loss=0.0378
|
| 107 |
+
step 1050000 | 9795s | train_loss=0.0379
|
| 108 |
+
step 1060000 | 9880s | train_loss=0.0376
|
| 109 |
+
step 1070000 | 9966s | train_loss=0.0372
|
| 110 |
+
step 1080000 | 10053s | train_loss=0.0363
|
| 111 |
+
step 1090000 | 10140s | train_loss=0.0365
|
| 112 |
+
step 1100000 | 10293s | train_loss=0.0372 | fid=193.7871
|
| 113 |
+
step 1110000 | 10380s | train_loss=0.0361
|
| 114 |
+
step 1120000 | 10467s | train_loss=0.0355
|
| 115 |
+
step 1130000 | 10553s | train_loss=0.0385
|
| 116 |
+
step 1140000 | 10640s | train_loss=0.0348
|
| 117 |
+
step 1150000 | 10727s | train_loss=0.0362
|
| 118 |
+
step 1160000 | 10812s | train_loss=0.0346
|
| 119 |
+
step 1170000 | 10898s | train_loss=0.0346
|
| 120 |
+
step 1180000 | 10985s | train_loss=0.0358
|
| 121 |
+
step 1190000 | 11071s | train_loss=0.0348
|
| 122 |
+
step 1200000 | 11223s | train_loss=0.0346 | fid=190.4956
|
| 123 |
+
step 1210000 | 11309s | train_loss=0.0345
|
| 124 |
+
step 1220000 | 11396s | train_loss=0.0351
|
| 125 |
+
step 1230000 | 11483s | train_loss=0.0347
|
| 126 |
+
step 1240000 | 11570s | train_loss=0.0339
|
| 127 |
+
step 1250000 | 11656s | train_loss=0.0342
|
| 128 |
+
step 1260000 | 11743s | train_loss=0.0354
|
| 129 |
+
step 1270000 | 11829s | train_loss=0.0331
|
| 130 |
+
step 1280000 | 11914s | train_loss=0.0336
|
| 131 |
+
step 1290000 | 12001s | train_loss=0.0349
|
| 132 |
+
step 1300000 | 12154s | train_loss=0.0338 | fid=188.1170
|
| 133 |
+
step 1310000 | 12241s | train_loss=0.0331
|
| 134 |
+
step 1320000 | 12328s | train_loss=0.0333
|
| 135 |
+
step 1330000 | 12415s | train_loss=0.0327
|
| 136 |
+
step 1340000 | 12502s | train_loss=0.0329
|
| 137 |
+
step 1350000 | 12588s | train_loss=0.0322
|
| 138 |
+
step 1360000 | 12675s | train_loss=0.0329
|
| 139 |
+
step 1370000 | 12761s | train_loss=0.0330
|
| 140 |
+
step 1380000 | 12846s | train_loss=0.0316
|
| 141 |
+
step 1390000 | 12933s | train_loss=0.0323
|
| 142 |
+
step 1400000 | 13087s | train_loss=0.0320 | fid=181.0860
|
| 143 |
+
step 1410000 | 13173s | train_loss=0.0330
|
| 144 |
+
step 1420000 | 13260s | train_loss=0.0329
|
| 145 |
+
step 1430000 | 13347s | train_loss=0.0313
|
| 146 |
+
step 1440000 | 13434s | train_loss=0.0312
|
| 147 |
+
step 1450000 | 13520s | train_loss=0.0314
|
| 148 |
+
step 1460000 | 13607s | train_loss=0.0313
|
| 149 |
+
step 1470000 | 13694s | train_loss=0.0307
|
| 150 |
+
step 1480000 | 13779s | train_loss=0.0325
|
| 151 |
+
step 1490000 | 13863s | train_loss=0.0317
|
| 152 |
+
step 1500000 | 14047s | train_loss=0.0319 | fid=170.5269
|
| 153 |
+
step 1510000 | 14134s | train_loss=0.0307
|
| 154 |
+
step 1520000 | 14220s | train_loss=0.0316
|
| 155 |
+
step 1530000 | 14307s | train_loss=0.0299
|
| 156 |
+
step 1540000 | 14394s | train_loss=0.0300
|
| 157 |
+
step 1550000 | 14481s | train_loss=0.0301
|
| 158 |
+
step 1560000 | 14567s | train_loss=0.0288
|
| 159 |
+
step 1570000 | 14654s | train_loss=0.0287
|
| 160 |
+
step 1580000 | 14740s | train_loss=0.0299
|
| 161 |
+
step 1590000 | 14823s | train_loss=0.0297
|
| 162 |
+
step 1600000 | 14979s | train_loss=0.0298 | fid=164.7147
|
| 163 |
+
step 1610000 | 15066s | train_loss=0.0294
|
| 164 |
+
step 1620000 | 15153s | train_loss=0.0303
|
| 165 |
+
step 1630000 | 15240s | train_loss=0.0305
|
| 166 |
+
step 1640000 | 15327s | train_loss=0.0285
|
| 167 |
+
step 1650000 | 15413s | train_loss=0.0281
|
| 168 |
+
step 1660000 | 15500s | train_loss=0.0280
|
| 169 |
+
step 1670000 | 15587s | train_loss=0.0283
|
| 170 |
+
step 1680000 | 15672s | train_loss=0.0280
|
| 171 |
+
step 1690000 | 15756s | train_loss=0.0291
|
| 172 |
+
step 1700000 | 15910s | train_loss=0.0290 | fid=154.3421
|
| 173 |
+
step 1710000 | 15997s | train_loss=0.0285
|
| 174 |
+
step 1720000 | 16083s | train_loss=0.0284
|
| 175 |
+
step 1730000 | 16170s | train_loss=0.0278
|
| 176 |
+
step 1740000 | 16257s | train_loss=0.0285
|
| 177 |
+
step 1750000 | 16344s | train_loss=0.0273
|
| 178 |
+
step 1760000 | 16431s | train_loss=0.0271
|
| 179 |
+
step 1770000 | 16517s | train_loss=0.0278
|
| 180 |
+
step 1780000 | 16604s | train_loss=0.0272
|
| 181 |
+
step 1790000 | 16686s | train_loss=0.0268
|
| 182 |
+
step 1800000 | 16879s | train_loss=0.0267 | fid=141.4906
|
| 183 |
+
step 1810000 | 16966s | train_loss=0.0279
|
| 184 |
+
step 1820000 | 17053s | train_loss=0.0265
|
| 185 |
+
step 1830000 | 17140s | train_loss=0.0277
|
| 186 |
+
step 1840000 | 17226s | train_loss=0.0270
|
| 187 |
+
step 1850000 | 17313s | train_loss=0.0264
|
| 188 |
+
step 1860000 | 17400s | train_loss=0.0259
|
| 189 |
+
step 1870000 | 17487s | train_loss=0.0263
|
| 190 |
+
step 1880000 | 17572s | train_loss=0.0253
|
| 191 |
+
step 1890000 | 17649s | train_loss=0.0270
|
| 192 |
+
step 1900000 | 17813s | train_loss=0.0264 | fid=136.2648
|
| 193 |
+
step 1910000 | 17899s | train_loss=0.0274
|
| 194 |
+
step 1920000 | 17986s | train_loss=0.0257
|
| 195 |
+
step 1930000 | 18073s | train_loss=0.0254
|
| 196 |
+
step 1940000 | 18160s | train_loss=0.0258
|
| 197 |
+
step 1950000 | 18247s | train_loss=0.0252
|
| 198 |
+
step 1960000 | 18333s | train_loss=0.0246
|
| 199 |
+
step 1970000 | 18419s | train_loss=0.0248
|
| 200 |
+
step 1980000 | 18505s | train_loss=0.0255
|
| 201 |
+
step 1990000 | 18588s | train_loss=0.0259
|
| 202 |
+
step 2000000 | 18744s | train_loss=0.0251 | fid=132.1927
|
| 203 |
+
step 2010000 | 18831s | train_loss=0.0245
|
| 204 |
+
step 2020000 | 18917s | train_loss=0.0251
|
| 205 |
+
step 2030000 | 19004s | train_loss=0.0242
|
| 206 |
+
step 2040000 | 19091s | train_loss=0.0244
|
| 207 |
+
step 2050000 | 19178s | train_loss=0.0253
|
| 208 |
+
step 2060000 | 19264s | train_loss=0.0243
|
| 209 |
+
step 2070000 | 19349s | train_loss=0.0245
|
| 210 |
+
step 2080000 | 19436s | train_loss=0.0252
|
| 211 |
+
step 2090000 | 19514s | train_loss=0.0246
|
| 212 |
+
step 2100000 | 19725s | train_loss=0.0249 | fid=125.9540
|
| 213 |
+
step 2110000 | 19812s | train_loss=0.0247
|
| 214 |
+
step 2120000 | 19899s | train_loss=0.0238
|
| 215 |
+
step 2130000 | 19986s | train_loss=0.0237
|
| 216 |
+
step 2140000 | 20073s | train_loss=0.0235
|
| 217 |
+
step 2150000 | 20159s | train_loss=0.0238
|
| 218 |
+
step 2160000 | 20244s | train_loss=0.0232
|
| 219 |
+
step 2170000 | 20331s | train_loss=0.0234
|
| 220 |
+
step 2180000 | 20417s | train_loss=0.0233
|
| 221 |
+
step 2190000 | 20491s | train_loss=0.0242
|
| 222 |
+
step 2200000 | 20690s | train_loss=0.0243 | fid=126.8428
|
| 223 |
+
step 2210000 | 20777s | train_loss=0.0231
|
| 224 |
+
step 2220000 | 20863s | train_loss=0.0230
|
| 225 |
+
step 2230000 | 20950s | train_loss=0.0228
|
| 226 |
+
step 2240000 | 21037s | train_loss=0.0234
|
| 227 |
+
step 2250000 | 21124s | train_loss=0.0234
|
| 228 |
+
step 2260000 | 21209s | train_loss=0.0236
|
| 229 |
+
step 2270000 | 21295s | train_loss=0.0228
|
| 230 |
+
step 2280000 | 21380s | train_loss=0.0220
|
| 231 |
+
step 2290000 | 21453s | train_loss=0.0227
|
| 232 |
+
step 2300000 | 21657s | train_loss=0.0220 | fid=126.5337
|
| 233 |
+
step 2310000 | 21743s | train_loss=0.0225
|
| 234 |
+
step 2320000 | 21830s | train_loss=0.0219
|
| 235 |
+
step 2330000 | 21917s | train_loss=0.0227
|
| 236 |
+
step 2340000 | 22004s | train_loss=0.0225
|
| 237 |
+
step 2350000 | 22088s | train_loss=0.0223
|
| 238 |
+
step 2360000 | 22175s | train_loss=0.0225
|
| 239 |
+
step 2370000 | 22262s | train_loss=0.0224
|
| 240 |
+
step 2380000 | 22345s | train_loss=0.0221
|
| 241 |
+
step 2390000 | 22421s | train_loss=0.0223
|
| 242 |
+
step 2400000 | 22627s | train_loss=0.0225 | fid=120.6658
|
| 243 |
+
step 2410000 | 22714s | train_loss=0.0221
|
| 244 |
+
step 2420000 | 22801s | train_loss=0.0224
|
| 245 |
+
step 2430000 | 22888s | train_loss=0.0214
|
| 246 |
+
step 2440000 | 22972s | train_loss=0.0214
|
| 247 |
+
step 2450000 | 23059s | train_loss=0.0216
|
| 248 |
+
step 2460000 | 23146s | train_loss=0.0224
|
| 249 |
+
step 2470000 | 23233s | train_loss=0.0216
|
| 250 |
+
step 2480000 | 23312s | train_loss=0.0227
|
| 251 |
+
step 2490000 | 23391s | train_loss=0.0215
|
| 252 |
+
step 2500000 | 23578s | train_loss=0.0211 | fid=122.3089
|
| 253 |
+
step 2510000 | 23665s | train_loss=0.0229
|
| 254 |
+
step 2520000 | 23752s | train_loss=0.0212
|
| 255 |
+
step 2530000 | 23837s | train_loss=0.0214
|
| 256 |
+
step 2540000 | 23924s | train_loss=0.0218
|
| 257 |
+
step 2550000 | 24010s | train_loss=0.0207
|
| 258 |
+
step 2560000 | 24097s | train_loss=0.0208
|
| 259 |
+
step 2570000 | 24184s | train_loss=0.0219
|
| 260 |
+
step 2580000 | 24265s | train_loss=0.0216
|
| 261 |
+
step 2590000 | 24340s | train_loss=0.0215
|
| 262 |
+
step 2600000 | 24538s | train_loss=0.0208 | fid=120.4403
|
| 263 |
+
step 2610000 | 24625s | train_loss=0.0208
|
| 264 |
+
step 2620000 | 24712s | train_loss=0.0216
|
| 265 |
+
step 2630000 | 24796s | train_loss=0.0218
|
| 266 |
+
step 2640000 | 24883s | train_loss=0.0203
|
| 267 |
+
step 2650000 | 24970s | train_loss=0.0202
|
| 268 |
+
step 2660000 | 25057s | train_loss=0.0212
|
| 269 |
+
step 2670000 | 25143s | train_loss=0.0202
|
| 270 |
+
step 2680000 | 25226s | train_loss=0.0207
|
| 271 |
+
step 2690000 | 25302s | train_loss=0.0207
|
| 272 |
+
step 2700000 | 25504s | train_loss=0.0201 | fid=120.4207
|
| 273 |
+
step 2710000 | 25591s | train_loss=0.0200
|
| 274 |
+
step 2720000 | 25676s | train_loss=0.0210
|
| 275 |
+
step 2730000 | 25762s | train_loss=0.0213
|
| 276 |
+
step 2740000 | 25849s | train_loss=0.0208
|
| 277 |
+
step 2750000 | 25936s | train_loss=0.0203
|
| 278 |
+
step 2760000 | 26023s | train_loss=0.0207
|
| 279 |
+
step 2770000 | 26109s | train_loss=0.0203
|
| 280 |
+
step 2780000 | 26190s | train_loss=0.0203
|
| 281 |
+
step 2790000 | 26270s | train_loss=0.0211
|
| 282 |
+
step 2800000 | 26432s | train_loss=0.0194 | fid=119.3159
|
| 283 |
+
step 2810000 | 26519s | train_loss=0.0195
|
| 284 |
+
step 2820000 | 26604s | train_loss=0.0196
|
| 285 |
+
step 2830000 | 26691s | train_loss=0.0189
|
| 286 |
+
step 2840000 | 26778s | train_loss=0.0193
|
| 287 |
+
step 2850000 | 26864s | train_loss=0.0198
|
| 288 |
+
step 2860000 | 26951s | train_loss=0.0205
|
| 289 |
+
step 2870000 | 27038s | train_loss=0.0201
|
| 290 |
+
step 2880000 | 27124s | train_loss=0.0190
|
| 291 |
+
step 2890000 | 27207s | train_loss=0.0200
|
| 292 |
+
step 2900000 | 27395s | train_loss=0.0194 | fid=121.3764
|
| 293 |
+
step 2910000 | 27480s | train_loss=0.0194
|
| 294 |
+
step 2920000 | 27566s | train_loss=0.0189
|
| 295 |
+
step 2930000 | 27653s | train_loss=0.0189
|
| 296 |
+
step 2940000 | 27740s | train_loss=0.0182
|
| 297 |
+
step 2950000 | 27827s | train_loss=0.0195
|
| 298 |
+
step 2960000 | 27914s | train_loss=0.0194
|
| 299 |
+
step 2970000 | 28000s | train_loss=0.0187
|
| 300 |
+
step 2980000 | 28077s | train_loss=0.0189
|
| 301 |
+
step 2990000 | 28160s | train_loss=0.0192
|
| 302 |
+
step 3000000 | 28343s | train_loss=0.0186 | fid=115.8858
|
| 303 |
+
step 3010000 | 28430s | train_loss=0.0187
|
| 304 |
+
step 3020000 | 28516s | train_loss=0.0180
|
| 305 |
+
step 3030000 | 28603s | train_loss=0.0185
|
| 306 |
+
step 3040000 | 28690s | train_loss=0.0192
|
| 307 |
+
step 3050000 | 28777s | train_loss=0.0187
|
| 308 |
+
step 3060000 | 28864s | train_loss=0.0184
|
| 309 |
+
step 3070000 | 28950s | train_loss=0.0189
|
| 310 |
+
step 3080000 | 29029s | train_loss=0.0183
|
| 311 |
+
step 3090000 | 29108s | train_loss=0.0187
|
| 312 |
+
step 3100000 | 29277s | train_loss=0.0187 | fid=116.6017
|
| 313 |
+
step 3110000 | 29362s | train_loss=0.0195
|
| 314 |
+
step 3120000 | 29449s | train_loss=0.0187
|
| 315 |
+
step 3130000 | 29536s | train_loss=0.0182
|
| 316 |
+
step 3140000 | 29622s | train_loss=0.0182
|
| 317 |
+
step 3150000 | 29709s | train_loss=0.0177
|
| 318 |
+
step 3160000 | 29796s | train_loss=0.0182
|
| 319 |
+
step 3170000 | 29883s | train_loss=0.0190
|
| 320 |
+
step 3180000 | 29969s | train_loss=0.0186
|
| 321 |
+
step 3190000 | 30054s | train_loss=0.0177
|
| 322 |
+
step 3200000 | 30209s | train_loss=0.0188 | fid=116.3636
|
| 323 |
+
step 3210000 | 30295s | train_loss=0.0183
|
| 324 |
+
step 3220000 | 30382s | train_loss=0.0177
|
| 325 |
+
step 3230000 | 30469s | train_loss=0.0187
|
| 326 |
+
step 3240000 | 30555s | train_loss=0.0179
|
| 327 |
+
step 3250000 | 30642s | train_loss=0.0185
|
| 328 |
+
step 3260000 | 30729s | train_loss=0.0211
|
| 329 |
+
step 3270000 | 30816s | train_loss=0.0171
|
| 330 |
+
step 3280000 | 30900s | train_loss=0.0181
|
| 331 |
+
step 3290000 | 30979s | train_loss=0.0164
|
| 332 |
+
step 3300000 | 31150s | train_loss=0.0179 | fid=115.2843
|
| 333 |
+
step 3310000 | 31237s | train_loss=0.0179
|
| 334 |
+
step 3320000 | 31322s | train_loss=0.0168
|
| 335 |
+
step 3330000 | 31409s | train_loss=0.0173
|
| 336 |
+
step 3340000 | 31496s | train_loss=0.0177
|
| 337 |
+
step 3350000 | 31583s | train_loss=0.0178
|
| 338 |
+
step 3360000 | 31670s | train_loss=0.0179
|
| 339 |
+
step 3370000 | 31756s | train_loss=0.0187
|
| 340 |
+
step 3380000 | 31843s | train_loss=0.0175
|
| 341 |
+
step 3390000 | 31920s | train_loss=0.0170
|
| 342 |
+
step 3400000 | 32100s | train_loss=0.0172 | fid=116.0041
|
| 343 |
+
step 3410000 | 32187s | train_loss=0.0170
|
| 344 |
+
step 3420000 | 32273s | train_loss=0.0172
|
| 345 |
+
step 3430000 | 32359s | train_loss=0.0177
|
| 346 |
+
step 3440000 | 32446s | train_loss=0.0174
|
| 347 |
+
step 3450000 | 32533s | train_loss=0.0174
|
| 348 |
+
step 3460000 | 32620s | train_loss=0.0176
|
| 349 |
+
step 3470000 | 32706s | train_loss=0.0170
|
| 350 |
+
step 3480000 | 32792s | train_loss=0.0172
|
| 351 |
+
step 3490000 | 32870s | train_loss=0.0177
|
| 352 |
+
step 3500000 | 33072s | train_loss=0.0170 | fid=116.2794
|
| 353 |
+
step 3510000 | 33159s | train_loss=0.0171
|
| 354 |
+
step 3520000 | 33245s | train_loss=0.0165
|
| 355 |
+
step 3530000 | 33331s | train_loss=0.0181
|
| 356 |
+
step 3540000 | 33418s | train_loss=0.0170
|
| 357 |
+
step 3550000 | 33505s | train_loss=0.0167
|
| 358 |
+
step 3560000 | 33592s | train_loss=0.0171
|
| 359 |
+
step 3570000 | 33679s | train_loss=0.0171
|
| 360 |
+
step 3580000 | 33758s | train_loss=0.0183
|
| 361 |
+
step 3590000 | 33841s | train_loss=0.0168
|
| 362 |
+
step 3600000 | 34020s | train_loss=0.0168 | fid=116.0628
|
| 363 |
+
step 3610000 | 34107s | train_loss=0.0171
|
| 364 |
+
step 3620000 | 34192s | train_loss=0.0166
|
| 365 |
+
step 3630000 | 34279s | train_loss=0.0167
|
| 366 |
+
step 3640000 | 34366s | train_loss=0.0174
|
| 367 |
+
step 3650000 | 34453s | train_loss=0.0182
|
| 368 |
+
step 3660000 | 34539s | train_loss=0.0158
|
| 369 |
+
step 3670000 | 34626s | train_loss=0.0174
|
| 370 |
+
step 3680000 | 34703s | train_loss=0.0170
|
| 371 |
+
step 3690000 | 34790s | train_loss=0.0168
|
| 372 |
+
step 3700000 | 34984s | train_loss=0.0160 | fid=113.3206
|
| 373 |
+
step 3710000 | 35071s | train_loss=0.0175
|
| 374 |
+
step 3720000 | 35156s | train_loss=0.0168
|
| 375 |
+
step 3730000 | 35243s | train_loss=0.0160
|
| 376 |
+
step 3740000 | 35330s | train_loss=0.0173
|
| 377 |
+
step 3750000 | 35416s | train_loss=0.0172
|
| 378 |
+
step 3760000 | 35503s | train_loss=0.0164
|
| 379 |
+
step 3770000 | 35590s | train_loss=0.0169
|
| 380 |
+
step 3780000 | 35659s | train_loss=0.0161
|
| 381 |
+
step 3790000 | 35746s | train_loss=0.0160
|
| 382 |
+
step 3800000 | 35940s | train_loss=0.0164 | fid=109.6248
|
| 383 |
+
step 3810000 | 36027s | train_loss=0.0160
|
| 384 |
+
step 3820000 | 36112s | train_loss=0.0163
|
| 385 |
+
step 3830000 | 36199s | train_loss=0.0161
|
| 386 |
+
step 3840000 | 36286s | train_loss=0.0165
|
| 387 |
+
step 3850000 | 36372s | train_loss=0.0171
|
| 388 |
+
step 3860000 | 36459s | train_loss=0.0156
|
| 389 |
+
step 3870000 | 36544s | train_loss=0.0178
|
| 390 |
+
step 3880000 | 36617s | train_loss=0.0155
|
| 391 |
+
step 3890000 | 36704s | train_loss=0.0163
|
| 392 |
+
step 3900000 | 36888s | train_loss=0.0161 | fid=111.8738
|
| 393 |
+
step 3910000 | 36975s | train_loss=0.0152
|
| 394 |
+
step 3920000 | 37060s | train_loss=0.0158
|
| 395 |
+
step 3930000 | 37147s | train_loss=0.0159
|
| 396 |
+
step 3940000 | 37234s | train_loss=0.0152
|
| 397 |
+
step 3950000 | 37321s | train_loss=0.0162
|
| 398 |
+
step 3960000 | 37408s | train_loss=0.0156
|
| 399 |
+
step 3970000 | 37492s | train_loss=0.0171
|
| 400 |
+
step 3980000 | 37569s | train_loss=0.0155
|
| 401 |
+
step 3990000 | 37650s | train_loss=0.0153
|
| 402 |
+
step 4000000 | 37840s | train_loss=0.0153 | fid=110.3164
|
| 403 |
+
step 4010000 | 37926s | train_loss=0.0156
|
| 404 |
+
step 4020000 | 38014s | train_loss=0.0160
|
| 405 |
+
step 4030000 | 38099s | train_loss=0.0160
|
| 406 |
+
step 4040000 | 38185s | train_loss=0.0148
|
| 407 |
+
step 4050000 | 38272s | train_loss=0.0162
|
| 408 |
+
step 4060000 | 38357s | train_loss=0.0164
|
| 409 |
+
step 4070000 | 38444s | train_loss=0.0152
|
| 410 |
+
step 4080000 | 38524s | train_loss=0.0152
|
| 411 |
+
step 4090000 | 38604s | train_loss=0.0146
|
| 412 |
+
step 4100000 | 38787s | train_loss=0.0148 | fid=111.0584
|
| 413 |
+
step 4110000 | 38873s | train_loss=0.0158
|
| 414 |
+
step 4120000 | 38960s | train_loss=0.0154
|
| 415 |
+
step 4130000 | 39046s | train_loss=0.0160
|
| 416 |
+
step 4140000 | 39133s | train_loss=0.0162
|
| 417 |
+
step 4150000 | 39218s | train_loss=0.0150
|
| 418 |
+
step 4160000 | 39305s | train_loss=0.0159
|
| 419 |
+
step 4170000 | 39391s | train_loss=0.0156
|
| 420 |
+
step 4180000 | 39474s | train_loss=0.0149
|
| 421 |
+
step 4190000 | 39550s | train_loss=0.0153
|
| 422 |
+
step 4200000 | 39746s | train_loss=0.0148 | fid=108.4177
|
| 423 |
+
step 4210000 | 39833s | train_loss=0.0146
|
| 424 |
+
step 4220000 | 39920s | train_loss=0.0154
|
| 425 |
+
step 4230000 | 40005s | train_loss=0.0150
|
| 426 |
+
step 4240000 | 40092s | train_loss=0.0160
|
| 427 |
+
step 4250000 | 40177s | train_loss=0.0148
|
| 428 |
+
step 4260000 | 40264s | train_loss=0.0151
|
| 429 |
+
step 4270000 | 40350s | train_loss=0.0144
|
| 430 |
+
step 4280000 | 40434s | train_loss=0.0157
|
| 431 |
+
step 4290000 | 40509s | train_loss=0.0152
|
| 432 |
+
step 4300000 | 40715s | train_loss=0.0150 | fid=108.5780
|
| 433 |
+
step 4310000 | 40802s | train_loss=0.0149
|
| 434 |
+
step 4320000 | 40889s | train_loss=0.0156
|
| 435 |
+
step 4330000 | 40974s | train_loss=0.0150
|
| 436 |
+
step 4340000 | 41059s | train_loss=0.0150
|
| 437 |
+
step 4350000 | 41145s | train_loss=0.0155
|
| 438 |
+
step 4360000 | 41232s | train_loss=0.0157
|
| 439 |
+
step 4370000 | 41319s | train_loss=0.0154
|
| 440 |
+
step 4380000 | 41401s | train_loss=0.0148
|
| 441 |
+
step 4390000 | 41478s | train_loss=0.0150
|
| 442 |
+
step 4400000 | 41679s | train_loss=0.0145 | fid=108.6456
|
| 443 |
+
step 4410000 | 41766s | train_loss=0.0154
|
| 444 |
+
step 4420000 | 41853s | train_loss=0.0148
|
| 445 |
+
step 4430000 | 41937s | train_loss=0.0144
|
| 446 |
+
step 4440000 | 42024s | train_loss=0.0145
|
| 447 |
+
step 4450000 | 42110s | train_loss=0.0152
|
| 448 |
+
step 4460000 | 42197s | train_loss=0.0141
|
| 449 |
+
step 4470000 | 42284s | train_loss=0.0151
|
| 450 |
+
step 4480000 | 42365s | train_loss=0.0138
|
| 451 |
+
step 4490000 | 42439s | train_loss=0.0146
|
| 452 |
+
step 4500000 | 42650s | train_loss=0.0143 | fid=107.9816
|
| 453 |
+
step 4510000 | 42737s | train_loss=0.0140
|
| 454 |
+
step 4520000 | 42820s | train_loss=0.0137
|
| 455 |
+
step 4530000 | 42907s | train_loss=0.0155
|
| 456 |
+
step 4540000 | 42994s | train_loss=0.0145
|
| 457 |
+
step 4550000 | 43081s | train_loss=0.0141
|
| 458 |
+
step 4560000 | 43167s | train_loss=0.0140
|
| 459 |
+
step 4570000 | 43254s | train_loss=0.0140
|
| 460 |
+
step 4580000 | 43335s | train_loss=0.0136
|
| 461 |
+
step 4590000 | 43420s | train_loss=0.0155
|
| 462 |
+
step 4600000 | 43618s | train_loss=0.0146 | fid=107.3112
|
| 463 |
+
step 4610000 | 43703s | train_loss=0.0143
|
| 464 |
+
step 4620000 | 43788s | train_loss=0.0150
|
| 465 |
+
step 4630000 | 43875s | train_loss=0.0149
|
| 466 |
+
step 4640000 | 43962s | train_loss=0.0148
|
| 467 |
+
step 4650000 | 44049s | train_loss=0.0157
|
| 468 |
+
step 4660000 | 44135s | train_loss=0.0143
|
| 469 |
+
step 4670000 | 44222s | train_loss=0.0139
|
| 470 |
+
step 4680000 | 44299s | train_loss=0.0138
|
| 471 |
+
step 4690000 | 44386s | train_loss=0.0147
|
| 472 |
+
step 4700000 | 44592s | train_loss=0.0144 | fid=107.3455
|
| 473 |
+
step 4710000 | 44678s | train_loss=0.0137
|
| 474 |
+
step 4720000 | 44763s | train_loss=0.0141
|
| 475 |
+
step 4730000 | 44850s | train_loss=0.0134
|
| 476 |
+
step 4740000 | 44937s | train_loss=0.0139
|
| 477 |
+
step 4750000 | 45024s | train_loss=0.0137
|
| 478 |
+
step 4760000 | 45110s | train_loss=0.0140
|
| 479 |
+
step 4770000 | 45195s | train_loss=0.0138
|
| 480 |
+
step 4780000 | 45271s | train_loss=0.0137
|
| 481 |
+
step 4790000 | 45358s | train_loss=0.0132
|
| 482 |
+
step 4800000 | 45538s | train_loss=0.0142 | fid=107.3229
|
| 483 |
+
step 4810000 | 45608s | train_loss=0.0146
|
| 484 |
+
step 4820000 | 45676s | train_loss=0.0140
|
| 485 |
+
step 4830000 | 45746s | train_loss=0.0137
|
| 486 |
+
step 4840000 | 45816s | train_loss=0.0133
|
| 487 |
+
step 4850000 | 45886s | train_loss=0.0130
|
| 488 |
+
step 4860000 | 45956s | train_loss=0.0135
|
| 489 |
+
step 4870000 | 46023s | train_loss=0.0132
|
| 490 |
+
step 4880000 | 46077s | train_loss=0.0132
|
| 491 |
+
step 4890000 | 46147s | train_loss=0.0143
|
| 492 |
+
step 4900000 | 46271s | train_loss=0.0138 | fid=108.2734
|
| 493 |
+
step 4910000 | 46341s | train_loss=0.0142
|
| 494 |
+
step 4920000 | 46411s | train_loss=0.0135
|
| 495 |
+
step 4930000 | 46479s | train_loss=0.0138
|
| 496 |
+
step 4940000 | 46549s | train_loss=0.0129
|
| 497 |
+
step 4950000 | 46619s | train_loss=0.0142
|
| 498 |
+
step 4960000 | 46689s | train_loss=0.0139
|
| 499 |
+
step 4970000 | 46759s | train_loss=0.0136
|
| 500 |
+
step 4980000 | 46824s | train_loss=0.0137
|
| 501 |
+
step 4990000 | 46877s | train_loss=0.0130
|
| 502 |
+
step 5000000 | 46971s | train_loss=0.0134 | fid=106.4819
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d120/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d120/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d120",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 45,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d120/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d120/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 68s | train_loss=1.0001 | fid=305.8222
|
| 3 |
+
step 10000 | 148s | train_loss=0.8116
|
| 4 |
+
step 20000 | 235s | train_loss=0.6875
|
| 5 |
+
step 30000 | 322s | train_loss=0.5600
|
| 6 |
+
step 40000 | 410s | train_loss=0.4362
|
| 7 |
+
step 50000 | 497s | train_loss=0.3322
|
| 8 |
+
step 60000 | 584s | train_loss=0.2562
|
| 9 |
+
step 70000 | 671s | train_loss=0.2028
|
| 10 |
+
step 80000 | 758s | train_loss=0.1710
|
| 11 |
+
step 90000 | 846s | train_loss=0.1496
|
| 12 |
+
step 100000 | 1000s | train_loss=0.1338 | fid=178.3425
|
| 13 |
+
step 110000 | 1083s | train_loss=0.1245
|
| 14 |
+
step 120000 | 1165s | train_loss=0.1179
|
| 15 |
+
step 130000 | 1252s | train_loss=0.1118
|
| 16 |
+
step 140000 | 1340s | train_loss=0.1046
|
| 17 |
+
step 150000 | 1427s | train_loss=0.1011
|
| 18 |
+
step 160000 | 1514s | train_loss=0.1005
|
| 19 |
+
step 170000 | 1602s | train_loss=0.0958
|
| 20 |
+
step 180000 | 1689s | train_loss=0.0914
|
| 21 |
+
step 190000 | 1776s | train_loss=0.0889
|
| 22 |
+
step 200000 | 1931s | train_loss=0.0880 | fid=176.7819
|
| 23 |
+
step 210000 | 2014s | train_loss=0.0842
|
| 24 |
+
step 220000 | 2102s | train_loss=0.0829
|
| 25 |
+
step 230000 | 2184s | train_loss=0.0795
|
| 26 |
+
step 240000 | 2272s | train_loss=0.0803
|
| 27 |
+
step 250000 | 2359s | train_loss=0.0797
|
| 28 |
+
step 260000 | 2446s | train_loss=0.0741
|
| 29 |
+
step 270000 | 2533s | train_loss=0.0747
|
| 30 |
+
step 280000 | 2621s | train_loss=0.0748
|
| 31 |
+
step 290000 | 2708s | train_loss=0.0732
|
| 32 |
+
step 300000 | 2862s | train_loss=0.0719 | fid=171.0588
|
| 33 |
+
step 310000 | 2944s | train_loss=0.0682
|
| 34 |
+
step 320000 | 3032s | train_loss=0.0681
|
| 35 |
+
step 330000 | 3119s | train_loss=0.0672
|
| 36 |
+
step 340000 | 3204s | train_loss=0.0676
|
| 37 |
+
step 350000 | 3290s | train_loss=0.0643
|
| 38 |
+
step 360000 | 3378s | train_loss=0.0637
|
| 39 |
+
step 370000 | 3465s | train_loss=0.0624
|
| 40 |
+
step 380000 | 3552s | train_loss=0.0637
|
| 41 |
+
step 390000 | 3639s | train_loss=0.0615
|
| 42 |
+
step 400000 | 3793s | train_loss=0.0628 | fid=166.6376
|
| 43 |
+
step 410000 | 3879s | train_loss=0.0615
|
| 44 |
+
step 420000 | 3963s | train_loss=0.0596
|
| 45 |
+
step 430000 | 4050s | train_loss=0.0599
|
| 46 |
+
step 440000 | 4138s | train_loss=0.0566
|
| 47 |
+
step 450000 | 4226s | train_loss=0.0586
|
| 48 |
+
step 460000 | 4309s | train_loss=0.0566
|
| 49 |
+
step 470000 | 4396s | train_loss=0.0574
|
| 50 |
+
step 480000 | 4483s | train_loss=0.0567
|
| 51 |
+
step 490000 | 4571s | train_loss=0.0555
|
| 52 |
+
step 500000 | 4726s | train_loss=0.0554 | fid=166.0214
|
| 53 |
+
step 510000 | 4813s | train_loss=0.0536
|
| 54 |
+
step 520000 | 4896s | train_loss=0.0559
|
| 55 |
+
step 530000 | 4983s | train_loss=0.0535
|
| 56 |
+
step 540000 | 5070s | train_loss=0.0532
|
| 57 |
+
step 550000 | 5157s | train_loss=0.0561
|
| 58 |
+
step 560000 | 5243s | train_loss=0.0527
|
| 59 |
+
step 570000 | 5324s | train_loss=0.0520
|
| 60 |
+
step 580000 | 5410s | train_loss=0.0527
|
| 61 |
+
step 590000 | 5496s | train_loss=0.0518
|
| 62 |
+
step 600000 | 5650s | train_loss=0.0517 | fid=164.4093
|
| 63 |
+
step 610000 | 5737s | train_loss=0.0505
|
| 64 |
+
step 620000 | 5820s | train_loss=0.0525
|
| 65 |
+
step 630000 | 5907s | train_loss=0.0510
|
| 66 |
+
step 640000 | 5994s | train_loss=0.0499
|
| 67 |
+
step 650000 | 6081s | train_loss=0.0488
|
| 68 |
+
step 660000 | 6168s | train_loss=0.0490
|
| 69 |
+
step 670000 | 6256s | train_loss=0.0514
|
| 70 |
+
step 680000 | 6338s | train_loss=0.0494
|
| 71 |
+
step 690000 | 6426s | train_loss=0.0479
|
| 72 |
+
step 700000 | 6581s | train_loss=0.0464 | fid=163.8211
|
| 73 |
+
step 710000 | 6668s | train_loss=0.0475
|
| 74 |
+
step 720000 | 6751s | train_loss=0.0477
|
| 75 |
+
step 730000 | 6838s | train_loss=0.0476
|
| 76 |
+
step 740000 | 6925s | train_loss=0.0484
|
| 77 |
+
step 750000 | 7013s | train_loss=0.0447
|
| 78 |
+
step 760000 | 7100s | train_loss=0.0454
|
| 79 |
+
step 770000 | 7187s | train_loss=0.0460
|
| 80 |
+
step 780000 | 7275s | train_loss=0.0466
|
| 81 |
+
step 790000 | 7360s | train_loss=0.0460
|
| 82 |
+
step 800000 | 7513s | train_loss=0.0451 | fid=175.8341
|
| 83 |
+
step 810000 | 7600s | train_loss=0.0435
|
| 84 |
+
step 820000 | 7685s | train_loss=0.0450
|
| 85 |
+
step 830000 | 7770s | train_loss=0.0446
|
| 86 |
+
step 840000 | 7857s | train_loss=0.0446
|
| 87 |
+
step 850000 | 7944s | train_loss=0.0462
|
| 88 |
+
step 860000 | 8031s | train_loss=0.0430
|
| 89 |
+
step 870000 | 8119s | train_loss=0.0444
|
| 90 |
+
step 880000 | 8206s | train_loss=0.0431
|
| 91 |
+
step 890000 | 8293s | train_loss=0.0422
|
| 92 |
+
step 900000 | 8443s | train_loss=0.0435 | fid=191.8173
|
| 93 |
+
step 910000 | 8530s | train_loss=0.0424
|
| 94 |
+
step 920000 | 8618s | train_loss=0.0433
|
| 95 |
+
step 930000 | 8701s | train_loss=0.0414
|
| 96 |
+
step 940000 | 8788s | train_loss=0.0416
|
| 97 |
+
step 950000 | 8875s | train_loss=0.0423
|
| 98 |
+
step 960000 | 8962s | train_loss=0.0413
|
| 99 |
+
step 970000 | 9050s | train_loss=0.0406
|
| 100 |
+
step 980000 | 9137s | train_loss=0.0407
|
| 101 |
+
step 990000 | 9224s | train_loss=0.0403
|
| 102 |
+
step 1000000 | 9379s | train_loss=0.0420 | fid=205.5440
|
| 103 |
+
step 1010000 | 9462s | train_loss=0.0414
|
| 104 |
+
step 1020000 | 9550s | train_loss=0.0408
|
| 105 |
+
step 1030000 | 9632s | train_loss=0.0408
|
| 106 |
+
step 1040000 | 9719s | train_loss=0.0418
|
| 107 |
+
step 1050000 | 9807s | train_loss=0.0399
|
| 108 |
+
step 1060000 | 9894s | train_loss=0.0424
|
| 109 |
+
step 1070000 | 9981s | train_loss=0.0413
|
| 110 |
+
step 1080000 | 10068s | train_loss=0.0394
|
| 111 |
+
step 1090000 | 10155s | train_loss=0.0382
|
| 112 |
+
step 1100000 | 10310s | train_loss=0.0386 | fid=210.0032
|
| 113 |
+
step 1110000 | 10397s | train_loss=0.0386
|
| 114 |
+
step 1120000 | 10482s | train_loss=0.0395
|
| 115 |
+
step 1130000 | 10563s | train_loss=0.0383
|
| 116 |
+
step 1140000 | 10649s | train_loss=0.0410
|
| 117 |
+
step 1150000 | 10736s | train_loss=0.0370
|
| 118 |
+
step 1160000 | 10824s | train_loss=0.0388
|
| 119 |
+
step 1170000 | 10911s | train_loss=0.0378
|
| 120 |
+
step 1180000 | 10998s | train_loss=0.0397
|
| 121 |
+
step 1190000 | 11085s | train_loss=0.0384
|
| 122 |
+
step 1200000 | 11240s | train_loss=0.0383 | fid=219.3611
|
| 123 |
+
step 1210000 | 11328s | train_loss=0.0366
|
| 124 |
+
step 1220000 | 11415s | train_loss=0.0375
|
| 125 |
+
step 1230000 | 11497s | train_loss=0.0369
|
| 126 |
+
step 1240000 | 11578s | train_loss=0.0382
|
| 127 |
+
step 1250000 | 11665s | train_loss=0.0370
|
| 128 |
+
step 1260000 | 11752s | train_loss=0.0381
|
| 129 |
+
step 1270000 | 11840s | train_loss=0.0367
|
| 130 |
+
step 1280000 | 11927s | train_loss=0.0363
|
| 131 |
+
step 1290000 | 12014s | train_loss=0.0357
|
| 132 |
+
step 1300000 | 12169s | train_loss=0.0361 | fid=226.5422
|
| 133 |
+
step 1310000 | 12256s | train_loss=0.0362
|
| 134 |
+
step 1320000 | 12344s | train_loss=0.0364
|
| 135 |
+
step 1330000 | 12431s | train_loss=0.0365
|
| 136 |
+
step 1340000 | 12514s | train_loss=0.0361
|
| 137 |
+
step 1350000 | 12595s | train_loss=0.0355
|
| 138 |
+
step 1360000 | 12682s | train_loss=0.0353
|
| 139 |
+
step 1370000 | 12769s | train_loss=0.0342
|
| 140 |
+
step 1380000 | 12857s | train_loss=0.0340
|
| 141 |
+
step 1390000 | 12944s | train_loss=0.0357
|
| 142 |
+
step 1400000 | 13099s | train_loss=0.0348 | fid=223.5758
|
| 143 |
+
step 1410000 | 13186s | train_loss=0.0363
|
| 144 |
+
step 1420000 | 13273s | train_loss=0.0353
|
| 145 |
+
step 1430000 | 13361s | train_loss=0.0340
|
| 146 |
+
step 1440000 | 13444s | train_loss=0.0358
|
| 147 |
+
step 1450000 | 13532s | train_loss=0.0344
|
| 148 |
+
step 1460000 | 13617s | train_loss=0.0339
|
| 149 |
+
step 1470000 | 13702s | train_loss=0.0354
|
| 150 |
+
step 1480000 | 13789s | train_loss=0.0351
|
| 151 |
+
step 1490000 | 13876s | train_loss=0.0347
|
| 152 |
+
step 1500000 | 14031s | train_loss=0.0329 | fid=218.7912
|
| 153 |
+
step 1510000 | 14118s | train_loss=0.0333
|
| 154 |
+
step 1520000 | 14206s | train_loss=0.0347
|
| 155 |
+
step 1530000 | 14293s | train_loss=0.0323
|
| 156 |
+
step 1540000 | 14379s | train_loss=0.0318
|
| 157 |
+
step 1550000 | 14453s | train_loss=0.0326
|
| 158 |
+
step 1560000 | 14541s | train_loss=0.0331
|
| 159 |
+
step 1570000 | 14622s | train_loss=0.0329
|
| 160 |
+
step 1580000 | 14700s | train_loss=0.0356
|
| 161 |
+
step 1590000 | 14788s | train_loss=0.0337
|
| 162 |
+
step 1600000 | 14943s | train_loss=0.0320 | fid=203.9432
|
| 163 |
+
step 1610000 | 15030s | train_loss=0.0314
|
| 164 |
+
step 1620000 | 15117s | train_loss=0.0318
|
| 165 |
+
step 1630000 | 15204s | train_loss=0.0338
|
| 166 |
+
step 1640000 | 15290s | train_loss=0.0325
|
| 167 |
+
step 1650000 | 15374s | train_loss=0.0316
|
| 168 |
+
step 1660000 | 15461s | train_loss=0.0316
|
| 169 |
+
step 1670000 | 15548s | train_loss=0.0311
|
| 170 |
+
step 1680000 | 15636s | train_loss=0.0303
|
| 171 |
+
step 1690000 | 15717s | train_loss=0.0320
|
| 172 |
+
step 1700000 | 15871s | train_loss=0.0306 | fid=193.0521
|
| 173 |
+
step 1710000 | 15958s | train_loss=0.0317
|
| 174 |
+
step 1720000 | 16046s | train_loss=0.0298
|
| 175 |
+
step 1730000 | 16133s | train_loss=0.0317
|
| 176 |
+
step 1740000 | 16220s | train_loss=0.0300
|
| 177 |
+
step 1750000 | 16306s | train_loss=0.0300
|
| 178 |
+
step 1760000 | 16382s | train_loss=0.0304
|
| 179 |
+
step 1770000 | 16469s | train_loss=0.0302
|
| 180 |
+
step 1780000 | 16556s | train_loss=0.0316
|
| 181 |
+
step 1790000 | 16644s | train_loss=0.0304
|
| 182 |
+
step 1800000 | 16782s | train_loss=0.0300 | fid=183.9751
|
| 183 |
+
step 1810000 | 16868s | train_loss=0.0298
|
| 184 |
+
step 1820000 | 16955s | train_loss=0.0307
|
| 185 |
+
step 1830000 | 17042s | train_loss=0.0290
|
| 186 |
+
step 1840000 | 17129s | train_loss=0.0285
|
| 187 |
+
step 1850000 | 17214s | train_loss=0.0302
|
| 188 |
+
step 1860000 | 17300s | train_loss=0.0305
|
| 189 |
+
step 1870000 | 17373s | train_loss=0.0299
|
| 190 |
+
step 1880000 | 17460s | train_loss=0.0293
|
| 191 |
+
step 1890000 | 17547s | train_loss=0.0294
|
| 192 |
+
step 1900000 | 17702s | train_loss=0.0279 | fid=173.9774
|
| 193 |
+
step 1910000 | 17780s | train_loss=0.0293
|
| 194 |
+
step 1920000 | 17860s | train_loss=0.0285
|
| 195 |
+
step 1930000 | 17947s | train_loss=0.0286
|
| 196 |
+
step 1940000 | 18034s | train_loss=0.0279
|
| 197 |
+
step 1950000 | 18120s | train_loss=0.0283
|
| 198 |
+
step 1960000 | 18207s | train_loss=0.0277
|
| 199 |
+
step 1970000 | 18294s | train_loss=0.0289
|
| 200 |
+
step 1980000 | 18379s | train_loss=0.0292
|
| 201 |
+
step 1990000 | 18466s | train_loss=0.0281
|
| 202 |
+
step 2000000 | 18621s | train_loss=0.0283 | fid=161.9242
|
| 203 |
+
step 2010000 | 18708s | train_loss=0.0269
|
| 204 |
+
step 2020000 | 18796s | train_loss=0.0270
|
| 205 |
+
step 2030000 | 18878s | train_loss=0.0269
|
| 206 |
+
step 2040000 | 18965s | train_loss=0.0278
|
| 207 |
+
step 2050000 | 19050s | train_loss=0.0271
|
| 208 |
+
step 2060000 | 19137s | train_loss=0.0282
|
| 209 |
+
step 2070000 | 19225s | train_loss=0.0270
|
| 210 |
+
step 2080000 | 19309s | train_loss=0.0278
|
| 211 |
+
step 2090000 | 19396s | train_loss=0.0269
|
| 212 |
+
step 2100000 | 19551s | train_loss=0.0280 | fid=157.6559
|
| 213 |
+
step 2110000 | 19638s | train_loss=0.0261
|
| 214 |
+
step 2120000 | 19726s | train_loss=0.0266
|
| 215 |
+
step 2130000 | 19813s | train_loss=0.0262
|
| 216 |
+
step 2140000 | 19893s | train_loss=0.0253
|
| 217 |
+
step 2150000 | 19954s | train_loss=0.0267
|
| 218 |
+
step 2160000 | 20042s | train_loss=0.0263
|
| 219 |
+
step 2170000 | 20129s | train_loss=0.0256
|
| 220 |
+
step 2180000 | 20208s | train_loss=0.0257
|
| 221 |
+
step 2190000 | 20292s | train_loss=0.0260
|
| 222 |
+
step 2200000 | 20447s | train_loss=0.0258 | fid=153.0719
|
| 223 |
+
step 2210000 | 20534s | train_loss=0.0248
|
| 224 |
+
step 2220000 | 20621s | train_loss=0.0251
|
| 225 |
+
step 2230000 | 20708s | train_loss=0.0244
|
| 226 |
+
step 2240000 | 20796s | train_loss=0.0258
|
| 227 |
+
step 2250000 | 20883s | train_loss=0.0260
|
| 228 |
+
step 2260000 | 20961s | train_loss=0.0257
|
| 229 |
+
step 2270000 | 21031s | train_loss=0.0242
|
| 230 |
+
step 2280000 | 21119s | train_loss=0.0256
|
| 231 |
+
step 2290000 | 21198s | train_loss=0.0260
|
| 232 |
+
step 2300000 | 21353s | train_loss=0.0247 | fid=150.9286
|
| 233 |
+
step 2310000 | 21440s | train_loss=0.0252
|
| 234 |
+
step 2320000 | 21527s | train_loss=0.0246
|
| 235 |
+
step 2330000 | 21615s | train_loss=0.0246
|
| 236 |
+
step 2340000 | 21702s | train_loss=0.0239
|
| 237 |
+
step 2350000 | 21789s | train_loss=0.0251
|
| 238 |
+
step 2360000 | 21875s | train_loss=0.0250
|
| 239 |
+
step 2370000 | 21962s | train_loss=0.0247
|
| 240 |
+
step 2380000 | 22036s | train_loss=0.0245
|
| 241 |
+
step 2390000 | 22109s | train_loss=0.0243
|
| 242 |
+
step 2400000 | 22250s | train_loss=0.0256 | fid=143.2544
|
| 243 |
+
step 2410000 | 22337s | train_loss=0.0252
|
| 244 |
+
step 2420000 | 22424s | train_loss=0.0242
|
| 245 |
+
step 2430000 | 22512s | train_loss=0.0243
|
| 246 |
+
step 2440000 | 22599s | train_loss=0.0234
|
| 247 |
+
step 2450000 | 22686s | train_loss=0.0238
|
| 248 |
+
step 2460000 | 22771s | train_loss=0.0238
|
| 249 |
+
step 2470000 | 22858s | train_loss=0.0242
|
| 250 |
+
step 2480000 | 22946s | train_loss=0.0254
|
| 251 |
+
step 2490000 | 23033s | train_loss=0.0245
|
| 252 |
+
step 2500000 | 23158s | train_loss=0.0228 | fid=138.6873
|
| 253 |
+
step 2510000 | 23245s | train_loss=0.0231
|
| 254 |
+
step 2520000 | 23332s | train_loss=0.0234
|
| 255 |
+
step 2530000 | 23420s | train_loss=0.0233
|
| 256 |
+
step 2540000 | 23507s | train_loss=0.0244
|
| 257 |
+
step 2550000 | 23594s | train_loss=0.0236
|
| 258 |
+
step 2560000 | 23680s | train_loss=0.0240
|
| 259 |
+
step 2570000 | 23767s | train_loss=0.0232
|
| 260 |
+
step 2580000 | 23855s | train_loss=0.0241
|
| 261 |
+
step 2590000 | 23942s | train_loss=0.0234
|
| 262 |
+
step 2600000 | 24097s | train_loss=0.0232 | fid=135.9454
|
| 263 |
+
step 2610000 | 24160s | train_loss=0.0228
|
| 264 |
+
step 2620000 | 24241s | train_loss=0.0232
|
| 265 |
+
step 2630000 | 24328s | train_loss=0.0240
|
| 266 |
+
step 2640000 | 24416s | train_loss=0.0232
|
| 267 |
+
step 2650000 | 24503s | train_loss=0.0225
|
| 268 |
+
step 2660000 | 24590s | train_loss=0.0222
|
| 269 |
+
step 2670000 | 24675s | train_loss=0.0226
|
| 270 |
+
step 2680000 | 24762s | train_loss=0.0238
|
| 271 |
+
step 2690000 | 24850s | train_loss=0.0235
|
| 272 |
+
step 2700000 | 25005s | train_loss=0.0223 | fid=134.7994
|
| 273 |
+
step 2710000 | 25087s | train_loss=0.0231
|
| 274 |
+
step 2720000 | 25158s | train_loss=0.0219
|
| 275 |
+
step 2730000 | 25229s | train_loss=0.0221
|
| 276 |
+
step 2740000 | 25315s | train_loss=0.0229
|
| 277 |
+
step 2750000 | 25402s | train_loss=0.0225
|
| 278 |
+
step 2760000 | 25490s | train_loss=0.0218
|
| 279 |
+
step 2770000 | 25575s | train_loss=0.0218
|
| 280 |
+
step 2780000 | 25663s | train_loss=0.0230
|
| 281 |
+
step 2790000 | 25750s | train_loss=0.0237
|
| 282 |
+
step 2800000 | 25905s | train_loss=0.0212 | fid=134.5446
|
| 283 |
+
step 2810000 | 25992s | train_loss=0.0223
|
| 284 |
+
step 2820000 | 26074s | train_loss=0.0213
|
| 285 |
+
step 2830000 | 26153s | train_loss=0.0210
|
| 286 |
+
step 2840000 | 26233s | train_loss=0.0219
|
| 287 |
+
step 2850000 | 26302s | train_loss=0.0213
|
| 288 |
+
step 2860000 | 26385s | train_loss=0.0209
|
| 289 |
+
step 2870000 | 26470s | train_loss=0.0224
|
| 290 |
+
step 2880000 | 26557s | train_loss=0.0220
|
| 291 |
+
step 2890000 | 26645s | train_loss=0.0220
|
| 292 |
+
step 2900000 | 26799s | train_loss=0.0217 | fid=129.3551
|
| 293 |
+
step 2910000 | 26886s | train_loss=0.0213
|
| 294 |
+
step 2920000 | 26973s | train_loss=0.0214
|
| 295 |
+
step 2930000 | 27057s | train_loss=0.0200
|
| 296 |
+
step 2940000 | 27145s | train_loss=0.0225
|
| 297 |
+
step 2950000 | 27232s | train_loss=0.0225
|
| 298 |
+
step 2960000 | 27308s | train_loss=0.0205
|
| 299 |
+
step 2970000 | 27379s | train_loss=0.0207
|
| 300 |
+
step 2980000 | 27466s | train_loss=0.0203
|
| 301 |
+
step 2990000 | 27554s | train_loss=0.0203
|
| 302 |
+
step 3000000 | 27709s | train_loss=0.0207 | fid=126.3363
|
| 303 |
+
step 3010000 | 27796s | train_loss=0.0209
|
| 304 |
+
step 3020000 | 27883s | train_loss=0.0199
|
| 305 |
+
step 3030000 | 27969s | train_loss=0.0211
|
| 306 |
+
step 3040000 | 28045s | train_loss=0.0214
|
| 307 |
+
step 3050000 | 28133s | train_loss=0.0204
|
| 308 |
+
step 3060000 | 28220s | train_loss=0.0216
|
| 309 |
+
step 3070000 | 28307s | train_loss=0.0207
|
| 310 |
+
step 3080000 | 28382s | train_loss=0.0203
|
| 311 |
+
step 3090000 | 28455s | train_loss=0.0200
|
| 312 |
+
step 3100000 | 28607s | train_loss=0.0213 | fid=124.3798
|
| 313 |
+
step 3110000 | 28695s | train_loss=0.0224
|
| 314 |
+
step 3120000 | 28782s | train_loss=0.0202
|
| 315 |
+
step 3130000 | 28869s | train_loss=0.0198
|
| 316 |
+
step 3140000 | 28952s | train_loss=0.0199
|
| 317 |
+
step 3150000 | 29029s | train_loss=0.0201
|
| 318 |
+
step 3160000 | 29117s | train_loss=0.0204
|
| 319 |
+
step 3170000 | 29204s | train_loss=0.0212
|
| 320 |
+
step 3180000 | 29289s | train_loss=0.0197
|
| 321 |
+
step 3190000 | 29376s | train_loss=0.0193
|
| 322 |
+
step 3200000 | 29510s | train_loss=0.0216 | fid=125.7977
|
| 323 |
+
step 3210000 | 29591s | train_loss=0.0192
|
| 324 |
+
step 3220000 | 29679s | train_loss=0.0204
|
| 325 |
+
step 3230000 | 29766s | train_loss=0.0207
|
| 326 |
+
step 3240000 | 29853s | train_loss=0.0198
|
| 327 |
+
step 3250000 | 29937s | train_loss=0.0212
|
| 328 |
+
step 3260000 | 30024s | train_loss=0.0216
|
| 329 |
+
step 3270000 | 30112s | train_loss=0.0198
|
| 330 |
+
step 3280000 | 30197s | train_loss=0.0194
|
| 331 |
+
step 3290000 | 30284s | train_loss=0.0183
|
| 332 |
+
step 3300000 | 30439s | train_loss=0.0194 | fid=124.5633
|
| 333 |
+
step 3310000 | 30523s | train_loss=0.0195
|
| 334 |
+
step 3320000 | 30608s | train_loss=0.0188
|
| 335 |
+
step 3330000 | 30695s | train_loss=0.0199
|
| 336 |
+
step 3340000 | 30782s | train_loss=0.0196
|
| 337 |
+
step 3350000 | 30869s | train_loss=0.0199
|
| 338 |
+
step 3360000 | 30954s | train_loss=0.0191
|
| 339 |
+
step 3370000 | 31041s | train_loss=0.0208
|
| 340 |
+
step 3380000 | 31127s | train_loss=0.0196
|
| 341 |
+
step 3390000 | 31214s | train_loss=0.0196
|
| 342 |
+
step 3400000 | 31369s | train_loss=0.0191 | fid=120.5332
|
| 343 |
+
step 3410000 | 31457s | train_loss=0.0195
|
| 344 |
+
step 3420000 | 31533s | train_loss=0.0198
|
| 345 |
+
step 3430000 | 31617s | train_loss=0.0192
|
| 346 |
+
step 3440000 | 31693s | train_loss=0.0196
|
| 347 |
+
step 3450000 | 31780s | train_loss=0.0194
|
| 348 |
+
step 3460000 | 31856s | train_loss=0.0195
|
| 349 |
+
step 3470000 | 31943s | train_loss=0.0203
|
| 350 |
+
step 3480000 | 32029s | train_loss=0.0189
|
| 351 |
+
step 3490000 | 32116s | train_loss=0.0188
|
| 352 |
+
step 3500000 | 32271s | train_loss=0.0192 | fid=123.1523
|
| 353 |
+
step 3510000 | 32358s | train_loss=0.0195
|
| 354 |
+
step 3520000 | 32446s | train_loss=0.0184
|
| 355 |
+
step 3530000 | 32533s | train_loss=0.0190
|
| 356 |
+
step 3540000 | 32606s | train_loss=0.0185
|
| 357 |
+
step 3550000 | 32690s | train_loss=0.0203
|
| 358 |
+
step 3560000 | 32769s | train_loss=0.0190
|
| 359 |
+
step 3570000 | 32846s | train_loss=0.0188
|
| 360 |
+
step 3580000 | 32933s | train_loss=0.0187
|
| 361 |
+
step 3590000 | 33018s | train_loss=0.0195
|
| 362 |
+
step 3600000 | 33172s | train_loss=0.0186 | fid=120.5396
|
| 363 |
+
step 3610000 | 33259s | train_loss=0.0187
|
| 364 |
+
step 3620000 | 33347s | train_loss=0.0184
|
| 365 |
+
step 3630000 | 33434s | train_loss=0.0187
|
| 366 |
+
step 3640000 | 33521s | train_loss=0.0188
|
| 367 |
+
step 3650000 | 33609s | train_loss=0.0190
|
| 368 |
+
step 3660000 | 33682s | train_loss=0.0183
|
| 369 |
+
step 3670000 | 33759s | train_loss=0.0187
|
| 370 |
+
step 3680000 | 33834s | train_loss=0.0191
|
| 371 |
+
step 3690000 | 33919s | train_loss=0.0189
|
| 372 |
+
step 3700000 | 34074s | train_loss=0.0191 | fid=116.2370
|
| 373 |
+
step 3710000 | 34161s | train_loss=0.0191
|
| 374 |
+
step 3720000 | 34248s | train_loss=0.0186
|
| 375 |
+
step 3730000 | 34336s | train_loss=0.0189
|
| 376 |
+
step 3740000 | 34423s | train_loss=0.0185
|
| 377 |
+
step 3750000 | 34510s | train_loss=0.0184
|
| 378 |
+
step 3760000 | 34598s | train_loss=0.0174
|
| 379 |
+
step 3770000 | 34685s | train_loss=0.0189
|
| 380 |
+
step 3780000 | 34755s | train_loss=0.0192
|
| 381 |
+
step 3790000 | 34823s | train_loss=0.0182
|
| 382 |
+
step 3800000 | 34976s | train_loss=0.0182 | fid=116.7505
|
| 383 |
+
step 3810000 | 35063s | train_loss=0.0177
|
| 384 |
+
step 3820000 | 35150s | train_loss=0.0180
|
| 385 |
+
step 3830000 | 35237s | train_loss=0.0181
|
| 386 |
+
step 3840000 | 35324s | train_loss=0.0178
|
| 387 |
+
step 3850000 | 35412s | train_loss=0.0185
|
| 388 |
+
step 3860000 | 35499s | train_loss=0.0171
|
| 389 |
+
step 3870000 | 35586s | train_loss=0.0184
|
| 390 |
+
step 3880000 | 35673s | train_loss=0.0178
|
| 391 |
+
step 3890000 | 35745s | train_loss=0.0191
|
| 392 |
+
step 3900000 | 35877s | train_loss=0.0175 | fid=117.5328
|
| 393 |
+
step 3910000 | 35964s | train_loss=0.0176
|
| 394 |
+
step 3920000 | 36051s | train_loss=0.0180
|
| 395 |
+
step 3930000 | 36139s | train_loss=0.0178
|
| 396 |
+
step 3940000 | 36226s | train_loss=0.0172
|
| 397 |
+
step 3950000 | 36313s | train_loss=0.0182
|
| 398 |
+
step 3960000 | 36401s | train_loss=0.0172
|
| 399 |
+
step 3970000 | 36488s | train_loss=0.0189
|
| 400 |
+
step 3980000 | 36575s | train_loss=0.0179
|
| 401 |
+
step 3990000 | 36662s | train_loss=0.0172
|
| 402 |
+
step 4000000 | 36801s | train_loss=0.0172 | fid=117.7019
|
| 403 |
+
step 4010000 | 36878s | train_loss=0.0188
|
| 404 |
+
step 4020000 | 36950s | train_loss=0.0188
|
| 405 |
+
step 4030000 | 37037s | train_loss=0.0183
|
| 406 |
+
step 4040000 | 37124s | train_loss=0.0174
|
| 407 |
+
step 4050000 | 37211s | train_loss=0.0175
|
| 408 |
+
step 4060000 | 37299s | train_loss=0.0179
|
| 409 |
+
step 4070000 | 37386s | train_loss=0.0180
|
| 410 |
+
step 4080000 | 37473s | train_loss=0.0169
|
| 411 |
+
step 4090000 | 37560s | train_loss=0.0167
|
| 412 |
+
step 4100000 | 37704s | train_loss=0.0175 | fid=115.9266
|
| 413 |
+
step 4110000 | 37792s | train_loss=0.0175
|
| 414 |
+
step 4120000 | 37874s | train_loss=0.0173
|
| 415 |
+
step 4130000 | 37949s | train_loss=0.0177
|
| 416 |
+
step 4140000 | 38021s | train_loss=0.0179
|
| 417 |
+
step 4150000 | 38107s | train_loss=0.0179
|
| 418 |
+
step 4160000 | 38195s | train_loss=0.0178
|
| 419 |
+
step 4170000 | 38282s | train_loss=0.0190
|
| 420 |
+
step 4180000 | 38369s | train_loss=0.0175
|
| 421 |
+
step 4190000 | 38456s | train_loss=0.0177
|
| 422 |
+
step 4200000 | 38609s | train_loss=0.0165 | fid=117.5819
|
| 423 |
+
step 4210000 | 38688s | train_loss=0.0172
|
| 424 |
+
step 4220000 | 38775s | train_loss=0.0175
|
| 425 |
+
step 4230000 | 38862s | train_loss=0.0166
|
| 426 |
+
step 4240000 | 38946s | train_loss=0.0170
|
| 427 |
+
step 4250000 | 39023s | train_loss=0.0174
|
| 428 |
+
step 4260000 | 39098s | train_loss=0.0172
|
| 429 |
+
step 4270000 | 39185s | train_loss=0.0171
|
| 430 |
+
step 4280000 | 39272s | train_loss=0.0175
|
| 431 |
+
step 4290000 | 39359s | train_loss=0.0166
|
| 432 |
+
step 4300000 | 39512s | train_loss=0.0175 | fid=115.6007
|
| 433 |
+
step 4310000 | 39594s | train_loss=0.0177
|
| 434 |
+
step 4320000 | 39678s | train_loss=0.0167
|
| 435 |
+
step 4330000 | 39765s | train_loss=0.0184
|
| 436 |
+
step 4340000 | 39852s | train_loss=0.0164
|
| 437 |
+
step 4350000 | 39939s | train_loss=0.0184
|
| 438 |
+
step 4360000 | 40022s | train_loss=0.0177
|
| 439 |
+
step 4370000 | 40102s | train_loss=0.0181
|
| 440 |
+
step 4380000 | 40179s | train_loss=0.0172
|
| 441 |
+
step 4390000 | 40266s | train_loss=0.0164
|
| 442 |
+
step 4400000 | 40420s | train_loss=0.0170 | fid=118.6264
|
| 443 |
+
step 4410000 | 40507s | train_loss=0.0167
|
| 444 |
+
step 4420000 | 40583s | train_loss=0.0170
|
| 445 |
+
step 4430000 | 40670s | train_loss=0.0164
|
| 446 |
+
step 4440000 | 40757s | train_loss=0.0167
|
| 447 |
+
step 4450000 | 40845s | train_loss=0.0168
|
| 448 |
+
step 4460000 | 40932s | train_loss=0.0162
|
| 449 |
+
step 4470000 | 41019s | train_loss=0.0163
|
| 450 |
+
step 4480000 | 41098s | train_loss=0.0158
|
| 451 |
+
step 4490000 | 41175s | train_loss=0.0164
|
| 452 |
+
step 4500000 | 41317s | train_loss=0.0162 | fid=115.8252
|
| 453 |
+
step 4510000 | 41405s | train_loss=0.0157
|
| 454 |
+
step 4520000 | 41492s | train_loss=0.0156
|
| 455 |
+
step 4530000 | 41567s | train_loss=0.0170
|
| 456 |
+
step 4540000 | 41654s | train_loss=0.0169
|
| 457 |
+
step 4550000 | 41742s | train_loss=0.0164
|
| 458 |
+
step 4560000 | 41829s | train_loss=0.0161
|
| 459 |
+
step 4570000 | 41916s | train_loss=0.0161
|
| 460 |
+
step 4580000 | 42003s | train_loss=0.0153
|
| 461 |
+
step 4590000 | 42091s | train_loss=0.0177
|
| 462 |
+
step 4600000 | 42230s | train_loss=0.0167 | fid=115.3232
|
| 463 |
+
step 4610000 | 42299s | train_loss=0.0162
|
| 464 |
+
step 4620000 | 42387s | train_loss=0.0157
|
| 465 |
+
step 4630000 | 42472s | train_loss=0.0172
|
| 466 |
+
step 4640000 | 42545s | train_loss=0.0162
|
| 467 |
+
step 4650000 | 42633s | train_loss=0.0157
|
| 468 |
+
step 4660000 | 42720s | train_loss=0.0159
|
| 469 |
+
step 4670000 | 42807s | train_loss=0.0167
|
| 470 |
+
step 4680000 | 42894s | train_loss=0.0160
|
| 471 |
+
step 4690000 | 42981s | train_loss=0.0163
|
| 472 |
+
step 4700000 | 43136s | train_loss=0.0163 | fid=113.7050
|
| 473 |
+
step 4710000 | 43214s | train_loss=0.0163
|
| 474 |
+
step 4720000 | 43293s | train_loss=0.0163
|
| 475 |
+
step 4730000 | 43372s | train_loss=0.0152
|
| 476 |
+
step 4740000 | 43458s | train_loss=0.0163
|
| 477 |
+
step 4750000 | 43537s | train_loss=0.0164
|
| 478 |
+
step 4760000 | 43624s | train_loss=0.0163
|
| 479 |
+
step 4770000 | 43711s | train_loss=0.0154
|
| 480 |
+
step 4780000 | 43798s | train_loss=0.0156
|
| 481 |
+
step 4790000 | 43886s | train_loss=0.0155
|
| 482 |
+
step 4800000 | 44041s | train_loss=0.0157 | fid=114.2620
|
| 483 |
+
step 4810000 | 44126s | train_loss=0.0154
|
| 484 |
+
step 4820000 | 44213s | train_loss=0.0170
|
| 485 |
+
step 4830000 | 44293s | train_loss=0.0156
|
| 486 |
+
step 4840000 | 44374s | train_loss=0.0154
|
| 487 |
+
step 4850000 | 44443s | train_loss=0.0150
|
| 488 |
+
step 4860000 | 44525s | train_loss=0.0158
|
| 489 |
+
step 4870000 | 44612s | train_loss=0.0151
|
| 490 |
+
step 4880000 | 44699s | train_loss=0.0152
|
| 491 |
+
step 4890000 | 44786s | train_loss=0.0160
|
| 492 |
+
step 4900000 | 44940s | train_loss=0.0168 | fid=116.8972
|
| 493 |
+
step 4910000 | 45026s | train_loss=0.0161
|
| 494 |
+
step 4920000 | 45113s | train_loss=0.0163
|
| 495 |
+
step 4930000 | 45200s | train_loss=0.0165
|
| 496 |
+
step 4940000 | 45288s | train_loss=0.0148
|
| 497 |
+
step 4950000 | 45363s | train_loss=0.0157
|
| 498 |
+
step 4960000 | 45435s | train_loss=0.0155
|
| 499 |
+
step 4970000 | 45513s | train_loss=0.0166
|
| 500 |
+
step 4980000 | 45601s | train_loss=0.0153
|
| 501 |
+
step 4990000 | 45688s | train_loss=0.0155
|
| 502 |
+
step 5000000 | 45843s | train_loss=0.0155 | fid=112.8595
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d140/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d140/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d140",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 45,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d140/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d140/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 72s | train_loss=0.9994 | fid=277.6460
|
| 3 |
+
step 10000 | 160s | train_loss=0.7943
|
| 4 |
+
step 20000 | 249s | train_loss=0.6552
|
| 5 |
+
step 30000 | 339s | train_loss=0.5264
|
| 6 |
+
step 40000 | 428s | train_loss=0.4143
|
| 7 |
+
step 50000 | 517s | train_loss=0.3217
|
| 8 |
+
step 60000 | 606s | train_loss=0.2547
|
| 9 |
+
step 70000 | 695s | train_loss=0.2046
|
| 10 |
+
step 80000 | 784s | train_loss=0.1713
|
| 11 |
+
step 90000 | 873s | train_loss=0.1490
|
| 12 |
+
step 100000 | 1030s | train_loss=0.1351 | fid=193.5986
|
| 13 |
+
step 110000 | 1115s | train_loss=0.1266
|
| 14 |
+
step 120000 | 1204s | train_loss=0.1184
|
| 15 |
+
step 130000 | 1293s | train_loss=0.1133
|
| 16 |
+
step 140000 | 1382s | train_loss=0.1096
|
| 17 |
+
step 150000 | 1471s | train_loss=0.1035
|
| 18 |
+
step 160000 | 1560s | train_loss=0.0964
|
| 19 |
+
step 170000 | 1649s | train_loss=0.0983
|
| 20 |
+
step 180000 | 1738s | train_loss=0.0927
|
| 21 |
+
step 190000 | 1827s | train_loss=0.0898
|
| 22 |
+
step 200000 | 1985s | train_loss=0.0884 | fid=201.1435
|
| 23 |
+
step 210000 | 2074s | train_loss=0.0887
|
| 24 |
+
step 220000 | 2160s | train_loss=0.0824
|
| 25 |
+
step 230000 | 2248s | train_loss=0.0836
|
| 26 |
+
step 240000 | 2337s | train_loss=0.0834
|
| 27 |
+
step 250000 | 2426s | train_loss=0.0786
|
| 28 |
+
step 260000 | 2515s | train_loss=0.0802
|
| 29 |
+
step 270000 | 2604s | train_loss=0.0773
|
| 30 |
+
step 280000 | 2693s | train_loss=0.0738
|
| 31 |
+
step 290000 | 2782s | train_loss=0.0726
|
| 32 |
+
step 300000 | 2940s | train_loss=0.0720 | fid=203.2141
|
| 33 |
+
step 310000 | 3029s | train_loss=0.0704
|
| 34 |
+
step 320000 | 3118s | train_loss=0.0696
|
| 35 |
+
step 330000 | 3205s | train_loss=0.0682
|
| 36 |
+
step 340000 | 3293s | train_loss=0.0688
|
| 37 |
+
step 350000 | 3382s | train_loss=0.0673
|
| 38 |
+
step 360000 | 3471s | train_loss=0.0665
|
| 39 |
+
step 370000 | 3560s | train_loss=0.0658
|
| 40 |
+
step 380000 | 3649s | train_loss=0.0639
|
| 41 |
+
step 390000 | 3738s | train_loss=0.0627
|
| 42 |
+
step 400000 | 3896s | train_loss=0.0653 | fid=191.9648
|
| 43 |
+
step 410000 | 3985s | train_loss=0.0621
|
| 44 |
+
step 420000 | 4074s | train_loss=0.0609
|
| 45 |
+
step 430000 | 4163s | train_loss=0.0615
|
| 46 |
+
step 440000 | 4249s | train_loss=0.0611
|
| 47 |
+
step 450000 | 4338s | train_loss=0.0609
|
| 48 |
+
step 460000 | 4427s | train_loss=0.0604
|
| 49 |
+
step 470000 | 4516s | train_loss=0.0607
|
| 50 |
+
step 480000 | 4605s | train_loss=0.0566
|
| 51 |
+
step 490000 | 4694s | train_loss=0.0566
|
| 52 |
+
step 500000 | 4851s | train_loss=0.0570 | fid=186.6369
|
| 53 |
+
step 510000 | 4940s | train_loss=0.0582
|
| 54 |
+
step 520000 | 5029s | train_loss=0.0575
|
| 55 |
+
step 530000 | 5118s | train_loss=0.0530
|
| 56 |
+
step 540000 | 5206s | train_loss=0.0522
|
| 57 |
+
step 550000 | 5289s | train_loss=0.0591
|
| 58 |
+
step 560000 | 5376s | train_loss=0.0537
|
| 59 |
+
step 570000 | 5465s | train_loss=0.0544
|
| 60 |
+
step 580000 | 5552s | train_loss=0.0534
|
| 61 |
+
step 590000 | 5641s | train_loss=0.0527
|
| 62 |
+
step 600000 | 5798s | train_loss=0.0550 | fid=179.9212
|
| 63 |
+
step 610000 | 5887s | train_loss=0.0515
|
| 64 |
+
step 620000 | 5976s | train_loss=0.0537
|
| 65 |
+
step 630000 | 6065s | train_loss=0.0511
|
| 66 |
+
step 640000 | 6154s | train_loss=0.0505
|
| 67 |
+
step 650000 | 6243s | train_loss=0.0524
|
| 68 |
+
step 660000 | 6328s | train_loss=0.0507
|
| 69 |
+
step 670000 | 6417s | train_loss=0.0537
|
| 70 |
+
step 680000 | 6506s | train_loss=0.0487
|
| 71 |
+
step 690000 | 6593s | train_loss=0.0473
|
| 72 |
+
step 700000 | 6752s | train_loss=0.0481 | fid=181.7180
|
| 73 |
+
step 710000 | 6841s | train_loss=0.0515
|
| 74 |
+
step 720000 | 6930s | train_loss=0.0485
|
| 75 |
+
step 730000 | 7019s | train_loss=0.0474
|
| 76 |
+
step 740000 | 7108s | train_loss=0.0499
|
| 77 |
+
step 750000 | 7197s | train_loss=0.0468
|
| 78 |
+
step 760000 | 7287s | train_loss=0.0488
|
| 79 |
+
step 770000 | 7372s | train_loss=0.0496
|
| 80 |
+
step 780000 | 7460s | train_loss=0.0480
|
| 81 |
+
step 790000 | 7547s | train_loss=0.0455
|
| 82 |
+
step 800000 | 7707s | train_loss=0.0463 | fid=190.6248
|
| 83 |
+
step 810000 | 7796s | train_loss=0.0451
|
| 84 |
+
step 820000 | 7885s | train_loss=0.0438
|
| 85 |
+
step 830000 | 7974s | train_loss=0.0468
|
| 86 |
+
step 840000 | 8063s | train_loss=0.0451
|
| 87 |
+
step 850000 | 8152s | train_loss=0.0450
|
| 88 |
+
step 860000 | 8241s | train_loss=0.0441
|
| 89 |
+
step 870000 | 8330s | train_loss=0.0436
|
| 90 |
+
step 880000 | 8415s | train_loss=0.0446
|
| 91 |
+
step 890000 | 8502s | train_loss=0.0456
|
| 92 |
+
step 900000 | 8661s | train_loss=0.0441 | fid=196.0288
|
| 93 |
+
step 910000 | 8750s | train_loss=0.0426
|
| 94 |
+
step 920000 | 8839s | train_loss=0.0448
|
| 95 |
+
step 930000 | 8928s | train_loss=0.0434
|
| 96 |
+
step 940000 | 9017s | train_loss=0.0437
|
| 97 |
+
step 950000 | 9106s | train_loss=0.0424
|
| 98 |
+
step 960000 | 9195s | train_loss=0.0408
|
| 99 |
+
step 970000 | 9284s | train_loss=0.0431
|
| 100 |
+
step 980000 | 9373s | train_loss=0.0417
|
| 101 |
+
step 990000 | 9457s | train_loss=0.0428
|
| 102 |
+
step 1000000 | 9615s | train_loss=0.0436 | fid=212.0041
|
| 103 |
+
step 1010000 | 9704s | train_loss=0.0423
|
| 104 |
+
step 1020000 | 9793s | train_loss=0.0424
|
| 105 |
+
step 1030000 | 9882s | train_loss=0.0442
|
| 106 |
+
step 1040000 | 9971s | train_loss=0.0450
|
| 107 |
+
step 1050000 | 10060s | train_loss=0.0409
|
| 108 |
+
step 1060000 | 10149s | train_loss=0.0392
|
| 109 |
+
step 1070000 | 10238s | train_loss=0.0419
|
| 110 |
+
step 1080000 | 10325s | train_loss=0.0422
|
| 111 |
+
step 1090000 | 10415s | train_loss=0.0397
|
| 112 |
+
step 1100000 | 10569s | train_loss=0.0398 | fid=225.2519
|
| 113 |
+
step 1110000 | 10659s | train_loss=0.0380
|
| 114 |
+
step 1120000 | 10748s | train_loss=0.0418
|
| 115 |
+
step 1130000 | 10837s | train_loss=0.0410
|
| 116 |
+
step 1140000 | 10926s | train_loss=0.0413
|
| 117 |
+
step 1150000 | 11015s | train_loss=0.0394
|
| 118 |
+
step 1160000 | 11104s | train_loss=0.0403
|
| 119 |
+
step 1170000 | 11193s | train_loss=0.0381
|
| 120 |
+
step 1180000 | 11280s | train_loss=0.0396
|
| 121 |
+
step 1190000 | 11369s | train_loss=0.0411
|
| 122 |
+
step 1200000 | 11526s | train_loss=0.0390 | fid=241.6185
|
| 123 |
+
step 1210000 | 11612s | train_loss=0.0376
|
| 124 |
+
step 1220000 | 11701s | train_loss=0.0387
|
| 125 |
+
step 1230000 | 11790s | train_loss=0.0380
|
| 126 |
+
step 1240000 | 11879s | train_loss=0.0394
|
| 127 |
+
step 1250000 | 11968s | train_loss=0.0396
|
| 128 |
+
step 1260000 | 12057s | train_loss=0.0389
|
| 129 |
+
step 1270000 | 12146s | train_loss=0.0381
|
| 130 |
+
step 1280000 | 12233s | train_loss=0.0371
|
| 131 |
+
step 1290000 | 12322s | train_loss=0.0371
|
| 132 |
+
step 1300000 | 12483s | train_loss=0.0380 | fid=240.3524
|
| 133 |
+
step 1310000 | 12570s | train_loss=0.0363
|
| 134 |
+
step 1320000 | 12655s | train_loss=0.0364
|
| 135 |
+
step 1330000 | 12744s | train_loss=0.0369
|
| 136 |
+
step 1340000 | 12833s | train_loss=0.0356
|
| 137 |
+
step 1350000 | 12922s | train_loss=0.0362
|
| 138 |
+
step 1360000 | 13011s | train_loss=0.0386
|
| 139 |
+
step 1370000 | 13099s | train_loss=0.0358
|
| 140 |
+
step 1380000 | 13187s | train_loss=0.0368
|
| 141 |
+
step 1390000 | 13277s | train_loss=0.0366
|
| 142 |
+
step 1400000 | 13433s | train_loss=0.0359 | fid=237.7237
|
| 143 |
+
step 1410000 | 13522s | train_loss=0.0361
|
| 144 |
+
step 1420000 | 13610s | train_loss=0.0364
|
| 145 |
+
step 1430000 | 13697s | train_loss=0.0350
|
| 146 |
+
step 1440000 | 13786s | train_loss=0.0377
|
| 147 |
+
step 1450000 | 13875s | train_loss=0.0384
|
| 148 |
+
step 1460000 | 13964s | train_loss=0.0357
|
| 149 |
+
step 1470000 | 14051s | train_loss=0.0355
|
| 150 |
+
step 1480000 | 14140s | train_loss=0.0354
|
| 151 |
+
step 1490000 | 14229s | train_loss=0.0345
|
| 152 |
+
step 1500000 | 14430s | train_loss=0.0333 | fid=237.4463
|
| 153 |
+
step 1510000 | 14519s | train_loss=0.0352
|
| 154 |
+
step 1520000 | 14606s | train_loss=0.0362
|
| 155 |
+
step 1530000 | 14682s | train_loss=0.0344
|
| 156 |
+
step 1540000 | 14771s | train_loss=0.0331
|
| 157 |
+
step 1550000 | 14860s | train_loss=0.0353
|
| 158 |
+
step 1560000 | 14947s | train_loss=0.0370
|
| 159 |
+
step 1570000 | 15036s | train_loss=0.0343
|
| 160 |
+
step 1580000 | 15125s | train_loss=0.0335
|
| 161 |
+
step 1590000 | 15214s | train_loss=0.0335
|
| 162 |
+
step 1600000 | 15375s | train_loss=0.0348 | fid=235.0503
|
| 163 |
+
step 1610000 | 15464s | train_loss=0.0334
|
| 164 |
+
step 1620000 | 15553s | train_loss=0.0327
|
| 165 |
+
step 1630000 | 15642s | train_loss=0.0323
|
| 166 |
+
step 1640000 | 15726s | train_loss=0.0324
|
| 167 |
+
step 1650000 | 15815s | train_loss=0.0338
|
| 168 |
+
step 1660000 | 15902s | train_loss=0.0322
|
| 169 |
+
step 1670000 | 15991s | train_loss=0.0313
|
| 170 |
+
step 1680000 | 16080s | train_loss=0.0326
|
| 171 |
+
step 1690000 | 16169s | train_loss=0.0314
|
| 172 |
+
step 1700000 | 16365s | train_loss=0.0331 | fid=218.6534
|
| 173 |
+
step 1710000 | 16454s | train_loss=0.0322
|
| 174 |
+
step 1720000 | 16543s | train_loss=0.0306
|
| 175 |
+
step 1730000 | 16632s | train_loss=0.0314
|
| 176 |
+
step 1740000 | 16718s | train_loss=0.0315
|
| 177 |
+
step 1750000 | 16787s | train_loss=0.0307
|
| 178 |
+
step 1760000 | 16876s | train_loss=0.0300
|
| 179 |
+
step 1770000 | 16965s | train_loss=0.0293
|
| 180 |
+
step 1780000 | 17054s | train_loss=0.0319
|
| 181 |
+
step 1790000 | 17143s | train_loss=0.0308
|
| 182 |
+
step 1800000 | 17359s | train_loss=0.0308 | fid=231.0242
|
| 183 |
+
step 1810000 | 17448s | train_loss=0.0332
|
| 184 |
+
step 1820000 | 17537s | train_loss=0.0326
|
| 185 |
+
step 1830000 | 17626s | train_loss=0.0307
|
| 186 |
+
step 1840000 | 17714s | train_loss=0.0288
|
| 187 |
+
step 1850000 | 17791s | train_loss=0.0338
|
| 188 |
+
step 1860000 | 17875s | train_loss=0.0297
|
| 189 |
+
step 1870000 | 17964s | train_loss=0.0309
|
| 190 |
+
step 1880000 | 18053s | train_loss=0.0298
|
| 191 |
+
step 1890000 | 18141s | train_loss=0.0320
|
| 192 |
+
step 1900000 | 18303s | train_loss=0.0294 | fid=220.9684
|
| 193 |
+
step 1910000 | 18392s | train_loss=0.0310
|
| 194 |
+
step 1920000 | 18481s | train_loss=0.0309
|
| 195 |
+
step 1930000 | 18570s | train_loss=0.0308
|
| 196 |
+
step 1940000 | 18657s | train_loss=0.0320
|
| 197 |
+
step 1950000 | 18746s | train_loss=0.0305
|
| 198 |
+
step 1960000 | 18833s | train_loss=0.0298
|
| 199 |
+
step 1970000 | 18919s | train_loss=0.0288
|
| 200 |
+
step 1980000 | 19008s | train_loss=0.0291
|
| 201 |
+
step 1990000 | 19095s | train_loss=0.0298
|
| 202 |
+
step 2000000 | 19258s | train_loss=0.0301 | fid=210.9921
|
| 203 |
+
step 2010000 | 19348s | train_loss=0.0274
|
| 204 |
+
step 2020000 | 19437s | train_loss=0.0309
|
| 205 |
+
step 2030000 | 19526s | train_loss=0.0277
|
| 206 |
+
step 2040000 | 19613s | train_loss=0.0271
|
| 207 |
+
step 2050000 | 19702s | train_loss=0.0279
|
| 208 |
+
step 2060000 | 19791s | train_loss=0.0303
|
| 209 |
+
step 2070000 | 19878s | train_loss=0.0283
|
| 210 |
+
step 2080000 | 19938s | train_loss=0.0291
|
| 211 |
+
step 2090000 | 20025s | train_loss=0.0288
|
| 212 |
+
step 2100000 | 20225s | train_loss=0.0300 | fid=207.1660
|
| 213 |
+
step 2110000 | 20314s | train_loss=0.0280
|
| 214 |
+
step 2120000 | 20403s | train_loss=0.0283
|
| 215 |
+
step 2130000 | 20490s | train_loss=0.0290
|
| 216 |
+
step 2140000 | 20579s | train_loss=0.0284
|
| 217 |
+
step 2150000 | 20668s | train_loss=0.0268
|
| 218 |
+
step 2160000 | 20757s | train_loss=0.0269
|
| 219 |
+
step 2170000 | 20846s | train_loss=0.0286
|
| 220 |
+
step 2180000 | 20934s | train_loss=0.0267
|
| 221 |
+
step 2190000 | 20999s | train_loss=0.0307
|
| 222 |
+
step 2200000 | 21183s | train_loss=0.0301 | fid=199.9686
|
| 223 |
+
step 2210000 | 21272s | train_loss=0.0255
|
| 224 |
+
step 2220000 | 21359s | train_loss=0.0269
|
| 225 |
+
step 2230000 | 21449s | train_loss=0.0264
|
| 226 |
+
step 2240000 | 21538s | train_loss=0.0275
|
| 227 |
+
step 2250000 | 21627s | train_loss=0.0282
|
| 228 |
+
step 2260000 | 21716s | train_loss=0.0272
|
| 229 |
+
step 2270000 | 21805s | train_loss=0.0261
|
| 230 |
+
step 2280000 | 21892s | train_loss=0.0262
|
| 231 |
+
step 2290000 | 21982s | train_loss=0.0267
|
| 232 |
+
step 2300000 | 22167s | train_loss=0.0268 | fid=197.4393
|
| 233 |
+
step 2310000 | 22254s | train_loss=0.0273
|
| 234 |
+
step 2320000 | 22343s | train_loss=0.0259
|
| 235 |
+
step 2330000 | 22432s | train_loss=0.0259
|
| 236 |
+
step 2340000 | 22521s | train_loss=0.0251
|
| 237 |
+
step 2350000 | 22611s | train_loss=0.0261
|
| 238 |
+
step 2360000 | 22700s | train_loss=0.0251
|
| 239 |
+
step 2370000 | 22787s | train_loss=0.0259
|
| 240 |
+
step 2380000 | 22876s | train_loss=0.0263
|
| 241 |
+
step 2390000 | 22965s | train_loss=0.0259
|
| 242 |
+
step 2400000 | 23153s | train_loss=0.0264 | fid=196.7680
|
| 243 |
+
step 2410000 | 23240s | train_loss=0.0267
|
| 244 |
+
step 2420000 | 23330s | train_loss=0.0246
|
| 245 |
+
step 2430000 | 23419s | train_loss=0.0254
|
| 246 |
+
step 2440000 | 23508s | train_loss=0.0248
|
| 247 |
+
step 2450000 | 23597s | train_loss=0.0238
|
| 248 |
+
step 2460000 | 23685s | train_loss=0.0262
|
| 249 |
+
step 2470000 | 23774s | train_loss=0.0252
|
| 250 |
+
step 2480000 | 23863s | train_loss=0.0242
|
| 251 |
+
step 2490000 | 23952s | train_loss=0.0254
|
| 252 |
+
step 2500000 | 24131s | train_loss=0.0241 | fid=183.0754
|
| 253 |
+
step 2510000 | 24208s | train_loss=0.0242
|
| 254 |
+
step 2520000 | 24297s | train_loss=0.0265
|
| 255 |
+
step 2530000 | 24386s | train_loss=0.0249
|
| 256 |
+
step 2540000 | 24475s | train_loss=0.0258
|
| 257 |
+
step 2550000 | 24564s | train_loss=0.0243
|
| 258 |
+
step 2560000 | 24651s | train_loss=0.0252
|
| 259 |
+
step 2570000 | 24740s | train_loss=0.0247
|
| 260 |
+
step 2580000 | 24829s | train_loss=0.0256
|
| 261 |
+
step 2590000 | 24918s | train_loss=0.0253
|
| 262 |
+
step 2600000 | 25127s | train_loss=0.0258 | fid=173.2385
|
| 263 |
+
step 2610000 | 25200s | train_loss=0.0253
|
| 264 |
+
step 2620000 | 25281s | train_loss=0.0243
|
| 265 |
+
step 2630000 | 25370s | train_loss=0.0235
|
| 266 |
+
step 2640000 | 25459s | train_loss=0.0222
|
| 267 |
+
step 2650000 | 25547s | train_loss=0.0244
|
| 268 |
+
step 2660000 | 25636s | train_loss=0.0231
|
| 269 |
+
step 2670000 | 25725s | train_loss=0.0235
|
| 270 |
+
step 2680000 | 25814s | train_loss=0.0259
|
| 271 |
+
step 2690000 | 25902s | train_loss=0.0230
|
| 272 |
+
step 2700000 | 26107s | train_loss=0.0241 | fid=178.4433
|
| 273 |
+
step 2710000 | 26197s | train_loss=0.0245
|
| 274 |
+
step 2720000 | 26268s | train_loss=0.0233
|
| 275 |
+
step 2730000 | 26344s | train_loss=0.0230
|
| 276 |
+
step 2740000 | 26432s | train_loss=0.0245
|
| 277 |
+
step 2750000 | 26521s | train_loss=0.0234
|
| 278 |
+
step 2760000 | 26610s | train_loss=0.0233
|
| 279 |
+
step 2770000 | 26699s | train_loss=0.0230
|
| 280 |
+
step 2780000 | 26788s | train_loss=0.0232
|
| 281 |
+
step 2790000 | 26875s | train_loss=0.0248
|
| 282 |
+
step 2800000 | 27038s | train_loss=0.0231 | fid=177.5987
|
| 283 |
+
step 2810000 | 27127s | train_loss=0.0223
|
| 284 |
+
step 2820000 | 27216s | train_loss=0.0230
|
| 285 |
+
step 2830000 | 27297s | train_loss=0.0231
|
| 286 |
+
step 2840000 | 27370s | train_loss=0.0229
|
| 287 |
+
step 2850000 | 27456s | train_loss=0.0224
|
| 288 |
+
step 2860000 | 27545s | train_loss=0.0229
|
| 289 |
+
step 2870000 | 27634s | train_loss=0.0220
|
| 290 |
+
step 2880000 | 27721s | train_loss=0.0239
|
| 291 |
+
step 2890000 | 27810s | train_loss=0.0236
|
| 292 |
+
step 2900000 | 28011s | train_loss=0.0227 | fid=175.2133
|
| 293 |
+
step 2910000 | 28100s | train_loss=0.0219
|
| 294 |
+
step 2920000 | 28189s | train_loss=0.0220
|
| 295 |
+
step 2930000 | 28278s | train_loss=0.0224
|
| 296 |
+
step 2940000 | 28358s | train_loss=0.0251
|
| 297 |
+
step 2950000 | 28435s | train_loss=0.0240
|
| 298 |
+
step 2960000 | 28517s | train_loss=0.0227
|
| 299 |
+
step 2970000 | 28605s | train_loss=0.0211
|
| 300 |
+
step 2980000 | 28694s | train_loss=0.0220
|
| 301 |
+
step 2990000 | 28783s | train_loss=0.0215
|
| 302 |
+
step 3000000 | 28992s | train_loss=0.0224 | fid=171.8919
|
| 303 |
+
step 3010000 | 29081s | train_loss=0.0225
|
| 304 |
+
step 3020000 | 29171s | train_loss=0.0211
|
| 305 |
+
step 3030000 | 29258s | train_loss=0.0218
|
| 306 |
+
step 3040000 | 29347s | train_loss=0.0233
|
| 307 |
+
step 3050000 | 29423s | train_loss=0.0225
|
| 308 |
+
step 3060000 | 29503s | train_loss=0.0237
|
| 309 |
+
step 3070000 | 29582s | train_loss=0.0228
|
| 310 |
+
step 3080000 | 29671s | train_loss=0.0238
|
| 311 |
+
step 3090000 | 29760s | train_loss=0.0214
|
| 312 |
+
step 3100000 | 29923s | train_loss=0.0210 | fid=165.1208
|
| 313 |
+
step 3110000 | 30012s | train_loss=0.0230
|
| 314 |
+
step 3120000 | 30101s | train_loss=0.0213
|
| 315 |
+
step 3130000 | 30189s | train_loss=0.0211
|
| 316 |
+
step 3140000 | 30278s | train_loss=0.0200
|
| 317 |
+
step 3150000 | 30367s | train_loss=0.0207
|
| 318 |
+
step 3160000 | 30453s | train_loss=0.0212
|
| 319 |
+
step 3170000 | 30541s | train_loss=0.0214
|
| 320 |
+
step 3180000 | 30627s | train_loss=0.0211
|
| 321 |
+
step 3190000 | 30716s | train_loss=0.0212
|
| 322 |
+
step 3200000 | 30878s | train_loss=0.0242 | fid=167.0856
|
| 323 |
+
step 3210000 | 30967s | train_loss=0.0216
|
| 324 |
+
step 3220000 | 31057s | train_loss=0.0213
|
| 325 |
+
step 3230000 | 31144s | train_loss=0.0215
|
| 326 |
+
step 3240000 | 31233s | train_loss=0.0223
|
| 327 |
+
step 3250000 | 31322s | train_loss=0.0237
|
| 328 |
+
step 3260000 | 31409s | train_loss=0.0212
|
| 329 |
+
step 3270000 | 31497s | train_loss=0.0211
|
| 330 |
+
step 3280000 | 31576s | train_loss=0.0207
|
| 331 |
+
step 3290000 | 31653s | train_loss=0.0202
|
| 332 |
+
step 3300000 | 31849s | train_loss=0.0213 | fid=159.6294
|
| 333 |
+
step 3310000 | 31938s | train_loss=0.0210
|
| 334 |
+
step 3320000 | 32026s | train_loss=0.0198
|
| 335 |
+
step 3330000 | 32114s | train_loss=0.0210
|
| 336 |
+
step 3340000 | 32203s | train_loss=0.0206
|
| 337 |
+
step 3350000 | 32291s | train_loss=0.0204
|
| 338 |
+
step 3360000 | 32380s | train_loss=0.0227
|
| 339 |
+
step 3370000 | 32469s | train_loss=0.0211
|
| 340 |
+
step 3380000 | 32556s | train_loss=0.0208
|
| 341 |
+
step 3390000 | 32632s | train_loss=0.0193
|
| 342 |
+
step 3400000 | 32816s | train_loss=0.0197 | fid=155.5277
|
| 343 |
+
step 3410000 | 32905s | train_loss=0.0211
|
| 344 |
+
step 3420000 | 32993s | train_loss=0.0213
|
| 345 |
+
step 3430000 | 33082s | train_loss=0.0196
|
| 346 |
+
step 3440000 | 33169s | train_loss=0.0222
|
| 347 |
+
step 3450000 | 33258s | train_loss=0.0204
|
| 348 |
+
step 3460000 | 33347s | train_loss=0.0206
|
| 349 |
+
step 3470000 | 33436s | train_loss=0.0219
|
| 350 |
+
step 3480000 | 33525s | train_loss=0.0203
|
| 351 |
+
step 3490000 | 33615s | train_loss=0.0199
|
| 352 |
+
step 3500000 | 33798s | train_loss=0.0204 | fid=151.1669
|
| 353 |
+
step 3510000 | 33885s | train_loss=0.0194
|
| 354 |
+
step 3520000 | 33974s | train_loss=0.0195
|
| 355 |
+
step 3530000 | 34063s | train_loss=0.0213
|
| 356 |
+
step 3540000 | 34150s | train_loss=0.0207
|
| 357 |
+
step 3550000 | 34239s | train_loss=0.0205
|
| 358 |
+
step 3560000 | 34328s | train_loss=0.0214
|
| 359 |
+
step 3570000 | 34417s | train_loss=0.0197
|
| 360 |
+
step 3580000 | 34507s | train_loss=0.0195
|
| 361 |
+
step 3590000 | 34596s | train_loss=0.0201
|
| 362 |
+
step 3600000 | 34774s | train_loss=0.0203 | fid=153.0366
|
| 363 |
+
step 3610000 | 34847s | train_loss=0.0210
|
| 364 |
+
step 3620000 | 34936s | train_loss=0.0205
|
| 365 |
+
step 3630000 | 35023s | train_loss=0.0204
|
| 366 |
+
step 3640000 | 35112s | train_loss=0.0198
|
| 367 |
+
step 3650000 | 35201s | train_loss=0.0204
|
| 368 |
+
step 3660000 | 35290s | train_loss=0.0191
|
| 369 |
+
step 3670000 | 35380s | train_loss=0.0207
|
| 370 |
+
step 3680000 | 35469s | train_loss=0.0194
|
| 371 |
+
step 3690000 | 35558s | train_loss=0.0195
|
| 372 |
+
step 3700000 | 35751s | train_loss=0.0202 | fid=148.4809
|
| 373 |
+
step 3710000 | 35832s | train_loss=0.0199
|
| 374 |
+
step 3720000 | 35908s | train_loss=0.0204
|
| 375 |
+
step 3730000 | 35998s | train_loss=0.0193
|
| 376 |
+
step 3740000 | 36087s | train_loss=0.0199
|
| 377 |
+
step 3750000 | 36176s | train_loss=0.0196
|
| 378 |
+
step 3760000 | 36265s | train_loss=0.0185
|
| 379 |
+
step 3770000 | 36354s | train_loss=0.0191
|
| 380 |
+
step 3780000 | 36443s | train_loss=0.0200
|
| 381 |
+
step 3790000 | 36532s | train_loss=0.0192
|
| 382 |
+
step 3800000 | 36740s | train_loss=0.0195 | fid=145.9279
|
| 383 |
+
step 3810000 | 36820s | train_loss=0.0196
|
| 384 |
+
step 3820000 | 36900s | train_loss=0.0183
|
| 385 |
+
step 3830000 | 36977s | train_loss=0.0192
|
| 386 |
+
step 3840000 | 37066s | train_loss=0.0197
|
| 387 |
+
step 3850000 | 37155s | train_loss=0.0203
|
| 388 |
+
step 3860000 | 37244s | train_loss=0.0183
|
| 389 |
+
step 3870000 | 37334s | train_loss=0.0211
|
| 390 |
+
step 3880000 | 37423s | train_loss=0.0206
|
| 391 |
+
step 3890000 | 37512s | train_loss=0.0219
|
| 392 |
+
step 3900000 | 37700s | train_loss=0.0187 | fid=150.1865
|
| 393 |
+
step 3910000 | 37788s | train_loss=0.0190
|
| 394 |
+
step 3920000 | 37873s | train_loss=0.0198
|
| 395 |
+
step 3930000 | 37949s | train_loss=0.0190
|
| 396 |
+
step 3940000 | 38022s | train_loss=0.0181
|
| 397 |
+
step 3950000 | 38111s | train_loss=0.0191
|
| 398 |
+
step 3960000 | 38200s | train_loss=0.0197
|
| 399 |
+
step 3970000 | 38289s | train_loss=0.0204
|
| 400 |
+
step 3980000 | 38378s | train_loss=0.0185
|
| 401 |
+
step 3990000 | 38467s | train_loss=0.0186
|
| 402 |
+
step 4000000 | 38649s | train_loss=0.0193 | fid=144.9035
|
| 403 |
+
step 4010000 | 38739s | train_loss=0.0189
|
| 404 |
+
step 4020000 | 38828s | train_loss=0.0205
|
| 405 |
+
step 4030000 | 38917s | train_loss=0.0198
|
| 406 |
+
step 4040000 | 38992s | train_loss=0.0188
|
| 407 |
+
step 4050000 | 39073s | train_loss=0.0185
|
| 408 |
+
step 4060000 | 39158s | train_loss=0.0186
|
| 409 |
+
step 4070000 | 39247s | train_loss=0.0188
|
| 410 |
+
step 4080000 | 39336s | train_loss=0.0181
|
| 411 |
+
step 4090000 | 39425s | train_loss=0.0180
|
| 412 |
+
step 4100000 | 39610s | train_loss=0.0177 | fid=136.3563
|
| 413 |
+
step 4110000 | 39699s | train_loss=0.0188
|
| 414 |
+
step 4120000 | 39788s | train_loss=0.0177
|
| 415 |
+
step 4130000 | 39877s | train_loss=0.0186
|
| 416 |
+
step 4140000 | 39966s | train_loss=0.0192
|
| 417 |
+
step 4150000 | 40044s | train_loss=0.0182
|
| 418 |
+
step 4160000 | 40127s | train_loss=0.0187
|
| 419 |
+
step 4170000 | 40212s | train_loss=0.0200
|
| 420 |
+
step 4180000 | 40301s | train_loss=0.0184
|
| 421 |
+
step 4190000 | 40388s | train_loss=0.0176
|
| 422 |
+
step 4200000 | 40582s | train_loss=0.0178 | fid=144.1735
|
| 423 |
+
step 4210000 | 40671s | train_loss=0.0189
|
| 424 |
+
step 4220000 | 40760s | train_loss=0.0183
|
| 425 |
+
step 4230000 | 40850s | train_loss=0.0169
|
| 426 |
+
step 4240000 | 40939s | train_loss=0.0181
|
| 427 |
+
step 4250000 | 41028s | train_loss=0.0184
|
| 428 |
+
step 4260000 | 41106s | train_loss=0.0182
|
| 429 |
+
step 4270000 | 41185s | train_loss=0.0184
|
| 430 |
+
step 4280000 | 41266s | train_loss=0.0181
|
| 431 |
+
step 4290000 | 41351s | train_loss=0.0189
|
| 432 |
+
step 4300000 | 41550s | train_loss=0.0179 | fid=137.4660
|
| 433 |
+
step 4310000 | 41639s | train_loss=0.0180
|
| 434 |
+
step 4320000 | 41728s | train_loss=0.0178
|
| 435 |
+
step 4330000 | 41818s | train_loss=0.0191
|
| 436 |
+
step 4340000 | 41907s | train_loss=0.0182
|
| 437 |
+
step 4350000 | 41996s | train_loss=0.0193
|
| 438 |
+
step 4360000 | 42085s | train_loss=0.0191
|
| 439 |
+
step 4370000 | 42165s | train_loss=0.0177
|
| 440 |
+
step 4380000 | 42240s | train_loss=0.0179
|
| 441 |
+
step 4390000 | 42314s | train_loss=0.0180
|
| 442 |
+
step 4400000 | 42528s | train_loss=0.0179 | fid=134.4000
|
| 443 |
+
step 4410000 | 42618s | train_loss=0.0190
|
| 444 |
+
step 4420000 | 42707s | train_loss=0.0179
|
| 445 |
+
step 4430000 | 42796s | train_loss=0.0178
|
| 446 |
+
step 4440000 | 42885s | train_loss=0.0182
|
| 447 |
+
step 4450000 | 42974s | train_loss=0.0174
|
| 448 |
+
step 4460000 | 43063s | train_loss=0.0168
|
| 449 |
+
step 4470000 | 43149s | train_loss=0.0175
|
| 450 |
+
step 4480000 | 43227s | train_loss=0.0173
|
| 451 |
+
step 4490000 | 43311s | train_loss=0.0172
|
| 452 |
+
step 4500000 | 43493s | train_loss=0.0167 | fid=132.2968
|
| 453 |
+
step 4510000 | 43582s | train_loss=0.0170
|
| 454 |
+
step 4520000 | 43671s | train_loss=0.0171
|
| 455 |
+
step 4530000 | 43761s | train_loss=0.0184
|
| 456 |
+
step 4540000 | 43850s | train_loss=0.0184
|
| 457 |
+
step 4550000 | 43939s | train_loss=0.0179
|
| 458 |
+
step 4560000 | 44027s | train_loss=0.0181
|
| 459 |
+
step 4570000 | 44113s | train_loss=0.0183
|
| 460 |
+
step 4580000 | 44202s | train_loss=0.0169
|
| 461 |
+
step 4590000 | 44286s | train_loss=0.0188
|
| 462 |
+
step 4600000 | 44463s | train_loss=0.0166 | fid=133.5960
|
| 463 |
+
step 4610000 | 44552s | train_loss=0.0172
|
| 464 |
+
step 4620000 | 44641s | train_loss=0.0184
|
| 465 |
+
step 4630000 | 44730s | train_loss=0.0192
|
| 466 |
+
step 4640000 | 44819s | train_loss=0.0176
|
| 467 |
+
step 4650000 | 44908s | train_loss=0.0190
|
| 468 |
+
step 4660000 | 44994s | train_loss=0.0189
|
| 469 |
+
step 4670000 | 45083s | train_loss=0.0168
|
| 470 |
+
step 4680000 | 45172s | train_loss=0.0176
|
| 471 |
+
step 4690000 | 45261s | train_loss=0.0181
|
| 472 |
+
step 4700000 | 45439s | train_loss=0.0171 | fid=129.5026
|
| 473 |
+
step 4710000 | 45521s | train_loss=0.0168
|
| 474 |
+
step 4720000 | 45610s | train_loss=0.0174
|
| 475 |
+
step 4730000 | 45699s | train_loss=0.0168
|
| 476 |
+
step 4740000 | 45788s | train_loss=0.0181
|
| 477 |
+
step 4750000 | 45869s | train_loss=0.0173
|
| 478 |
+
step 4760000 | 45939s | train_loss=0.0181
|
| 479 |
+
step 4770000 | 46011s | train_loss=0.0167
|
| 480 |
+
step 4780000 | 46082s | train_loss=0.0163
|
| 481 |
+
step 4790000 | 46154s | train_loss=0.0173
|
| 482 |
+
step 4800000 | 46309s | train_loss=0.0166 | fid=123.0053
|
| 483 |
+
step 4810000 | 46381s | train_loss=0.0182
|
| 484 |
+
step 4820000 | 46453s | train_loss=0.0178
|
| 485 |
+
step 4830000 | 46524s | train_loss=0.0165
|
| 486 |
+
step 4840000 | 46596s | train_loss=0.0166
|
| 487 |
+
step 4850000 | 46661s | train_loss=0.0161
|
| 488 |
+
step 4860000 | 46716s | train_loss=0.0173
|
| 489 |
+
step 4870000 | 46771s | train_loss=0.0165
|
| 490 |
+
step 4880000 | 46825s | train_loss=0.0164
|
| 491 |
+
step 4890000 | 46880s | train_loss=0.0176
|
| 492 |
+
step 4900000 | 46980s | train_loss=0.0181 | fid=126.2173
|
| 493 |
+
step 4910000 | 47031s | train_loss=0.0177
|
| 494 |
+
step 4920000 | 47077s | train_loss=0.0171
|
| 495 |
+
step 4930000 | 47132s | train_loss=0.0183
|
| 496 |
+
step 4940000 | 47186s | train_loss=0.0158
|
| 497 |
+
step 4950000 | 47241s | train_loss=0.0175
|
| 498 |
+
step 4960000 | 47295s | train_loss=0.0168
|
| 499 |
+
step 4970000 | 47350s | train_loss=0.0175
|
| 500 |
+
step 4980000 | 47404s | train_loss=0.0164
|
| 501 |
+
step 4990000 | 47459s | train_loss=0.0166
|
| 502 |
+
step 5000000 | 47561s | train_loss=0.0167 | fid=124.7373
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d160/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d160/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d160",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 45,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d160/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d160/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 69s | train_loss=1.0002 | fid=310.1915
|
| 3 |
+
step 10000 | 153s | train_loss=0.7869
|
| 4 |
+
step 20000 | 242s | train_loss=0.6696
|
| 5 |
+
step 30000 | 332s | train_loss=0.5521
|
| 6 |
+
step 40000 | 421s | train_loss=0.4520
|
| 7 |
+
step 50000 | 510s | train_loss=0.3690
|
| 8 |
+
step 60000 | 600s | train_loss=0.2993
|
| 9 |
+
step 70000 | 689s | train_loss=0.2423
|
| 10 |
+
step 80000 | 778s | train_loss=0.2170
|
| 11 |
+
step 90000 | 867s | train_loss=0.1789
|
| 12 |
+
step 100000 | 1023s | train_loss=0.1609 | fid=206.3334
|
| 13 |
+
step 110000 | 1110s | train_loss=0.1512
|
| 14 |
+
step 120000 | 1195s | train_loss=0.1362
|
| 15 |
+
step 130000 | 1284s | train_loss=0.1369
|
| 16 |
+
step 140000 | 1373s | train_loss=0.1268
|
| 17 |
+
step 150000 | 1463s | train_loss=0.1142
|
| 18 |
+
step 160000 | 1552s | train_loss=0.1110
|
| 19 |
+
step 170000 | 1641s | train_loss=0.1086
|
| 20 |
+
step 180000 | 1730s | train_loss=0.0986
|
| 21 |
+
step 190000 | 1820s | train_loss=0.0969
|
| 22 |
+
step 200000 | 1974s | train_loss=0.0988 | fid=274.6794
|
| 23 |
+
step 210000 | 2061s | train_loss=0.0971
|
| 24 |
+
step 220000 | 2151s | train_loss=0.0913
|
| 25 |
+
step 230000 | 2235s | train_loss=0.0912
|
| 26 |
+
step 240000 | 2325s | train_loss=0.0938
|
| 27 |
+
step 250000 | 2414s | train_loss=0.0836
|
| 28 |
+
step 260000 | 2503s | train_loss=0.0901
|
| 29 |
+
step 270000 | 2593s | train_loss=0.0818
|
| 30 |
+
step 280000 | 2682s | train_loss=0.0795
|
| 31 |
+
step 290000 | 2771s | train_loss=0.0800
|
| 32 |
+
step 300000 | 2927s | train_loss=0.0890 | fid=308.6878
|
| 33 |
+
step 310000 | 3014s | train_loss=0.0735
|
| 34 |
+
step 320000 | 3103s | train_loss=0.0754
|
| 35 |
+
step 330000 | 3193s | train_loss=0.0722
|
| 36 |
+
step 340000 | 3278s | train_loss=0.0738
|
| 37 |
+
step 350000 | 3368s | train_loss=0.0815
|
| 38 |
+
step 360000 | 3457s | train_loss=0.0717
|
| 39 |
+
step 370000 | 3546s | train_loss=0.0681
|
| 40 |
+
step 380000 | 3635s | train_loss=0.0660
|
| 41 |
+
step 390000 | 3725s | train_loss=0.0660
|
| 42 |
+
step 400000 | 3880s | train_loss=0.0691 | fid=327.4597
|
| 43 |
+
step 410000 | 3967s | train_loss=0.0672
|
| 44 |
+
step 420000 | 4056s | train_loss=0.0618
|
| 45 |
+
step 430000 | 4146s | train_loss=0.0661
|
| 46 |
+
step 440000 | 4235s | train_loss=0.0651
|
| 47 |
+
step 450000 | 4321s | train_loss=0.0657
|
| 48 |
+
step 460000 | 4411s | train_loss=0.0643
|
| 49 |
+
step 470000 | 4500s | train_loss=0.0677
|
| 50 |
+
step 480000 | 4589s | train_loss=0.0590
|
| 51 |
+
step 490000 | 4679s | train_loss=0.0578
|
| 52 |
+
step 500000 | 4834s | train_loss=0.0662 | fid=323.3066
|
| 53 |
+
step 510000 | 4921s | train_loss=0.0616
|
| 54 |
+
step 520000 | 5010s | train_loss=0.0588
|
| 55 |
+
step 530000 | 5099s | train_loss=0.0547
|
| 56 |
+
step 540000 | 5187s | train_loss=0.0560
|
| 57 |
+
step 550000 | 5274s | train_loss=0.0614
|
| 58 |
+
step 560000 | 5358s | train_loss=0.0547
|
| 59 |
+
step 570000 | 5446s | train_loss=0.0566
|
| 60 |
+
step 580000 | 5534s | train_loss=0.0573
|
| 61 |
+
step 590000 | 5622s | train_loss=0.0551
|
| 62 |
+
step 600000 | 5777s | train_loss=0.0531 | fid=308.3538
|
| 63 |
+
step 610000 | 5863s | train_loss=0.0541
|
| 64 |
+
step 620000 | 5953s | train_loss=0.0546
|
| 65 |
+
step 630000 | 6042s | train_loss=0.0529
|
| 66 |
+
step 640000 | 6131s | train_loss=0.0526
|
| 67 |
+
step 650000 | 6221s | train_loss=0.0558
|
| 68 |
+
step 660000 | 6309s | train_loss=0.0544
|
| 69 |
+
step 670000 | 6395s | train_loss=0.0536
|
| 70 |
+
step 680000 | 6485s | train_loss=0.0497
|
| 71 |
+
step 690000 | 6574s | train_loss=0.0497
|
| 72 |
+
step 700000 | 6730s | train_loss=0.0501 | fid=303.5888
|
| 73 |
+
step 710000 | 6816s | train_loss=0.0500
|
| 74 |
+
step 720000 | 6905s | train_loss=0.0476
|
| 75 |
+
step 730000 | 6995s | train_loss=0.0473
|
| 76 |
+
step 740000 | 7084s | train_loss=0.0502
|
| 77 |
+
step 750000 | 7173s | train_loss=0.0472
|
| 78 |
+
step 760000 | 7262s | train_loss=0.0475
|
| 79 |
+
step 770000 | 7350s | train_loss=0.0501
|
| 80 |
+
step 780000 | 7437s | train_loss=0.0453
|
| 81 |
+
step 790000 | 7524s | train_loss=0.0441
|
| 82 |
+
step 800000 | 7680s | train_loss=0.0481 | fid=288.4347
|
| 83 |
+
step 810000 | 7767s | train_loss=0.0434
|
| 84 |
+
step 820000 | 7856s | train_loss=0.0473
|
| 85 |
+
step 830000 | 7945s | train_loss=0.0481
|
| 86 |
+
step 840000 | 8035s | train_loss=0.0461
|
| 87 |
+
step 850000 | 8124s | train_loss=0.0452
|
| 88 |
+
step 860000 | 8213s | train_loss=0.0441
|
| 89 |
+
step 870000 | 8303s | train_loss=0.0441
|
| 90 |
+
step 880000 | 8390s | train_loss=0.0452
|
| 91 |
+
step 890000 | 8475s | train_loss=0.0438
|
| 92 |
+
step 900000 | 8631s | train_loss=0.0438 | fid=282.7847
|
| 93 |
+
step 910000 | 8718s | train_loss=0.0448
|
| 94 |
+
step 920000 | 8807s | train_loss=0.0430
|
| 95 |
+
step 930000 | 8896s | train_loss=0.0397
|
| 96 |
+
step 940000 | 8985s | train_loss=0.0448
|
| 97 |
+
step 950000 | 9074s | train_loss=0.0404
|
| 98 |
+
step 960000 | 9164s | train_loss=0.0433
|
| 99 |
+
step 970000 | 9253s | train_loss=0.0441
|
| 100 |
+
step 980000 | 9342s | train_loss=0.0415
|
| 101 |
+
step 990000 | 9427s | train_loss=0.0395
|
| 102 |
+
step 1000000 | 9583s | train_loss=0.0420 | fid=258.8830
|
| 103 |
+
step 1010000 | 9670s | train_loss=0.0443
|
| 104 |
+
step 1020000 | 9759s | train_loss=0.0413
|
| 105 |
+
step 1030000 | 9848s | train_loss=0.0436
|
| 106 |
+
step 1040000 | 9937s | train_loss=0.0408
|
| 107 |
+
step 1050000 | 10027s | train_loss=0.0393
|
| 108 |
+
step 1060000 | 10116s | train_loss=0.0437
|
| 109 |
+
step 1070000 | 10205s | train_loss=0.0393
|
| 110 |
+
step 1080000 | 10294s | train_loss=0.0406
|
| 111 |
+
step 1090000 | 10382s | train_loss=0.0366
|
| 112 |
+
step 1100000 | 10535s | train_loss=0.0379 | fid=232.0972
|
| 113 |
+
step 1110000 | 10621s | train_loss=0.0375
|
| 114 |
+
step 1120000 | 10710s | train_loss=0.0411
|
| 115 |
+
step 1130000 | 10799s | train_loss=0.0404
|
| 116 |
+
step 1140000 | 10888s | train_loss=0.0394
|
| 117 |
+
step 1150000 | 10978s | train_loss=0.0409
|
| 118 |
+
step 1160000 | 11067s | train_loss=0.0417
|
| 119 |
+
step 1170000 | 11156s | train_loss=0.0375
|
| 120 |
+
step 1180000 | 11243s | train_loss=0.0384
|
| 121 |
+
step 1190000 | 11332s | train_loss=0.0404
|
| 122 |
+
step 1200000 | 11490s | train_loss=0.0368 | fid=236.2950
|
| 123 |
+
step 1210000 | 11572s | train_loss=0.0365
|
| 124 |
+
step 1220000 | 11661s | train_loss=0.0365
|
| 125 |
+
step 1230000 | 11750s | train_loss=0.0379
|
| 126 |
+
step 1240000 | 11839s | train_loss=0.0388
|
| 127 |
+
step 1250000 | 11929s | train_loss=0.0393
|
| 128 |
+
step 1260000 | 12018s | train_loss=0.0413
|
| 129 |
+
step 1270000 | 12107s | train_loss=0.0365
|
| 130 |
+
step 1280000 | 12195s | train_loss=0.0349
|
| 131 |
+
step 1290000 | 12284s | train_loss=0.0368
|
| 132 |
+
step 1300000 | 12442s | train_loss=0.0354 | fid=225.0827
|
| 133 |
+
step 1310000 | 12528s | train_loss=0.0345
|
| 134 |
+
step 1320000 | 12612s | train_loss=0.0346
|
| 135 |
+
step 1330000 | 12701s | train_loss=0.0339
|
| 136 |
+
step 1340000 | 12790s | train_loss=0.0354
|
| 137 |
+
step 1350000 | 12879s | train_loss=0.0344
|
| 138 |
+
step 1360000 | 12968s | train_loss=0.0347
|
| 139 |
+
step 1370000 | 13057s | train_loss=0.0337
|
| 140 |
+
step 1380000 | 13145s | train_loss=0.0335
|
| 141 |
+
step 1390000 | 13234s | train_loss=0.0335
|
| 142 |
+
step 1400000 | 13390s | train_loss=0.0332 | fid=208.1587
|
| 143 |
+
step 1410000 | 13478s | train_loss=0.0355
|
| 144 |
+
step 1420000 | 13568s | train_loss=0.0356
|
| 145 |
+
step 1430000 | 13652s | train_loss=0.0317
|
| 146 |
+
step 1440000 | 13742s | train_loss=0.0333
|
| 147 |
+
step 1450000 | 13831s | train_loss=0.0350
|
| 148 |
+
step 1460000 | 13920s | train_loss=0.0319
|
| 149 |
+
step 1470000 | 14009s | train_loss=0.0313
|
| 150 |
+
step 1480000 | 14097s | train_loss=0.0352
|
| 151 |
+
step 1490000 | 14186s | train_loss=0.0324
|
| 152 |
+
step 1500000 | 14343s | train_loss=0.0322 | fid=211.3551
|
| 153 |
+
step 1510000 | 14421s | train_loss=0.0323
|
| 154 |
+
step 1520000 | 14509s | train_loss=0.0341
|
| 155 |
+
step 1530000 | 14598s | train_loss=0.0318
|
| 156 |
+
step 1540000 | 14673s | train_loss=0.0306
|
| 157 |
+
step 1550000 | 14762s | train_loss=0.0333
|
| 158 |
+
step 1560000 | 14851s | train_loss=0.0364
|
| 159 |
+
step 1570000 | 14939s | train_loss=0.0303
|
| 160 |
+
step 1580000 | 15027s | train_loss=0.0338
|
| 161 |
+
step 1590000 | 15117s | train_loss=0.0320
|
| 162 |
+
step 1600000 | 15274s | train_loss=0.0301 | fid=198.6044
|
| 163 |
+
step 1610000 | 15363s | train_loss=0.0305
|
| 164 |
+
step 1620000 | 15450s | train_loss=0.0331
|
| 165 |
+
step 1630000 | 15539s | train_loss=0.0344
|
| 166 |
+
step 1640000 | 15628s | train_loss=0.0312
|
| 167 |
+
step 1650000 | 15714s | train_loss=0.0321
|
| 168 |
+
step 1660000 | 15802s | train_loss=0.0304
|
| 169 |
+
step 1670000 | 15889s | train_loss=0.0312
|
| 170 |
+
step 1680000 | 15979s | train_loss=0.0297
|
| 171 |
+
step 1690000 | 16068s | train_loss=0.0303
|
| 172 |
+
step 1700000 | 16225s | train_loss=0.0305 | fid=190.6811
|
| 173 |
+
step 1710000 | 16314s | train_loss=0.0300
|
| 174 |
+
step 1720000 | 16392s | train_loss=0.0291
|
| 175 |
+
step 1730000 | 16481s | train_loss=0.0304
|
| 176 |
+
step 1740000 | 16570s | train_loss=0.0304
|
| 177 |
+
step 1750000 | 16659s | train_loss=0.0281
|
| 178 |
+
step 1760000 | 16740s | train_loss=0.0284
|
| 179 |
+
step 1770000 | 16814s | train_loss=0.0317
|
| 180 |
+
step 1780000 | 16903s | train_loss=0.0304
|
| 181 |
+
step 1790000 | 16992s | train_loss=0.0274
|
| 182 |
+
step 1800000 | 17150s | train_loss=0.0302 | fid=199.5987
|
| 183 |
+
step 1810000 | 17240s | train_loss=0.0330
|
| 184 |
+
step 1820000 | 17321s | train_loss=0.0310
|
| 185 |
+
step 1830000 | 17402s | train_loss=0.0287
|
| 186 |
+
step 1840000 | 17491s | train_loss=0.0269
|
| 187 |
+
step 1850000 | 17580s | train_loss=0.0306
|
| 188 |
+
step 1860000 | 17669s | train_loss=0.0276
|
| 189 |
+
step 1870000 | 17754s | train_loss=0.0282
|
| 190 |
+
step 1880000 | 17830s | train_loss=0.0278
|
| 191 |
+
step 1890000 | 17919s | train_loss=0.0289
|
| 192 |
+
step 1900000 | 18077s | train_loss=0.0287 | fid=195.0335
|
| 193 |
+
step 1910000 | 18166s | train_loss=0.0283
|
| 194 |
+
step 1920000 | 18255s | train_loss=0.0274
|
| 195 |
+
step 1930000 | 18342s | train_loss=0.0299
|
| 196 |
+
step 1940000 | 18431s | train_loss=0.0280
|
| 197 |
+
step 1950000 | 18520s | train_loss=0.0287
|
| 198 |
+
step 1960000 | 18609s | train_loss=0.0252
|
| 199 |
+
step 1970000 | 18697s | train_loss=0.0297
|
| 200 |
+
step 1980000 | 18786s | train_loss=0.0296
|
| 201 |
+
step 1990000 | 18870s | train_loss=0.0284
|
| 202 |
+
step 2000000 | 19027s | train_loss=0.0271 | fid=192.4477
|
| 203 |
+
step 2010000 | 19116s | train_loss=0.0274
|
| 204 |
+
step 2020000 | 19205s | train_loss=0.0277
|
| 205 |
+
step 2030000 | 19291s | train_loss=0.0285
|
| 206 |
+
step 2040000 | 19380s | train_loss=0.0273
|
| 207 |
+
step 2050000 | 19470s | train_loss=0.0276
|
| 208 |
+
step 2060000 | 19557s | train_loss=0.0276
|
| 209 |
+
step 2070000 | 19646s | train_loss=0.0288
|
| 210 |
+
step 2080000 | 19736s | train_loss=0.0274
|
| 211 |
+
step 2090000 | 19825s | train_loss=0.0280
|
| 212 |
+
step 2100000 | 19952s | train_loss=0.0265 | fid=187.4960
|
| 213 |
+
step 2110000 | 20041s | train_loss=0.0277
|
| 214 |
+
step 2120000 | 20130s | train_loss=0.0257
|
| 215 |
+
step 2130000 | 20211s | train_loss=0.0279
|
| 216 |
+
step 2140000 | 20296s | train_loss=0.0255
|
| 217 |
+
step 2150000 | 20386s | train_loss=0.0258
|
| 218 |
+
step 2160000 | 20473s | train_loss=0.0252
|
| 219 |
+
step 2170000 | 20562s | train_loss=0.0252
|
| 220 |
+
step 2180000 | 20651s | train_loss=0.0245
|
| 221 |
+
step 2190000 | 20740s | train_loss=0.0293
|
| 222 |
+
step 2200000 | 20899s | train_loss=0.0340 | fid=178.1385
|
| 223 |
+
step 2210000 | 20973s | train_loss=0.0252
|
| 224 |
+
step 2220000 | 21053s | train_loss=0.0245
|
| 225 |
+
step 2230000 | 21143s | train_loss=0.0246
|
| 226 |
+
step 2240000 | 21224s | train_loss=0.0264
|
| 227 |
+
step 2250000 | 21313s | train_loss=0.0278
|
| 228 |
+
step 2260000 | 21400s | train_loss=0.0262
|
| 229 |
+
step 2270000 | 21490s | train_loss=0.0256
|
| 230 |
+
step 2280000 | 21579s | train_loss=0.0263
|
| 231 |
+
step 2290000 | 21668s | train_loss=0.0251
|
| 232 |
+
step 2300000 | 21826s | train_loss=0.0264 | fid=171.3012
|
| 233 |
+
step 2310000 | 21915s | train_loss=0.0260
|
| 234 |
+
step 2320000 | 22004s | train_loss=0.0262
|
| 235 |
+
step 2330000 | 22067s | train_loss=0.0249
|
| 236 |
+
step 2340000 | 22146s | train_loss=0.0241
|
| 237 |
+
step 2350000 | 22230s | train_loss=0.0254
|
| 238 |
+
step 2360000 | 22318s | train_loss=0.0238
|
| 239 |
+
step 2370000 | 22407s | train_loss=0.0248
|
| 240 |
+
step 2380000 | 22496s | train_loss=0.0258
|
| 241 |
+
step 2390000 | 22585s | train_loss=0.0249
|
| 242 |
+
step 2400000 | 22743s | train_loss=0.0249 | fid=164.6875
|
| 243 |
+
step 2410000 | 22832s | train_loss=0.0258
|
| 244 |
+
step 2420000 | 22922s | train_loss=0.0245
|
| 245 |
+
step 2430000 | 23011s | train_loss=0.0240
|
| 246 |
+
step 2440000 | 23092s | train_loss=0.0249
|
| 247 |
+
step 2450000 | 23157s | train_loss=0.0239
|
| 248 |
+
step 2460000 | 23246s | train_loss=0.0244
|
| 249 |
+
step 2470000 | 23335s | train_loss=0.0244
|
| 250 |
+
step 2480000 | 23424s | train_loss=0.0270
|
| 251 |
+
step 2490000 | 23514s | train_loss=0.0240
|
| 252 |
+
step 2500000 | 23670s | train_loss=0.0245 | fid=168.7589
|
| 253 |
+
step 2510000 | 23760s | train_loss=0.0252
|
| 254 |
+
step 2520000 | 23849s | train_loss=0.0237
|
| 255 |
+
step 2530000 | 23938s | train_loss=0.0237
|
| 256 |
+
step 2540000 | 24027s | train_loss=0.0238
|
| 257 |
+
step 2550000 | 24109s | train_loss=0.0232
|
| 258 |
+
step 2560000 | 24175s | train_loss=0.0238
|
| 259 |
+
step 2570000 | 24263s | train_loss=0.0241
|
| 260 |
+
step 2580000 | 24353s | train_loss=0.0247
|
| 261 |
+
step 2590000 | 24442s | train_loss=0.0235
|
| 262 |
+
step 2600000 | 24599s | train_loss=0.0236 | fid=171.1880
|
| 263 |
+
step 2610000 | 24689s | train_loss=0.0225
|
| 264 |
+
step 2620000 | 24778s | train_loss=0.0238
|
| 265 |
+
step 2630000 | 24867s | train_loss=0.0261
|
| 266 |
+
step 2640000 | 24957s | train_loss=0.0226
|
| 267 |
+
step 2650000 | 25044s | train_loss=0.0243
|
| 268 |
+
step 2660000 | 25121s | train_loss=0.0233
|
| 269 |
+
step 2670000 | 25194s | train_loss=0.0249
|
| 270 |
+
step 2680000 | 25274s | train_loss=0.0274
|
| 271 |
+
step 2690000 | 25363s | train_loss=0.0238
|
| 272 |
+
step 2700000 | 25520s | train_loss=0.0225 | fid=164.1814
|
| 273 |
+
step 2710000 | 25609s | train_loss=0.0224
|
| 274 |
+
step 2720000 | 25698s | train_loss=0.0238
|
| 275 |
+
step 2730000 | 25787s | train_loss=0.0248
|
| 276 |
+
step 2740000 | 25876s | train_loss=0.0252
|
| 277 |
+
step 2750000 | 25964s | train_loss=0.0243
|
| 278 |
+
step 2760000 | 26053s | train_loss=0.0234
|
| 279 |
+
step 2770000 | 26129s | train_loss=0.0234
|
| 280 |
+
step 2780000 | 26215s | train_loss=0.0255
|
| 281 |
+
step 2790000 | 26286s | train_loss=0.0270
|
| 282 |
+
step 2800000 | 26435s | train_loss=0.0232 | fid=166.7125
|
| 283 |
+
step 2810000 | 26524s | train_loss=0.0241
|
| 284 |
+
step 2820000 | 26614s | train_loss=0.0224
|
| 285 |
+
step 2830000 | 26703s | train_loss=0.0220
|
| 286 |
+
step 2840000 | 26792s | train_loss=0.0232
|
| 287 |
+
step 2850000 | 26880s | train_loss=0.0226
|
| 288 |
+
step 2860000 | 26969s | train_loss=0.0252
|
| 289 |
+
step 2870000 | 27055s | train_loss=0.0233
|
| 290 |
+
step 2880000 | 27145s | train_loss=0.0237
|
| 291 |
+
step 2890000 | 27234s | train_loss=0.0254
|
| 292 |
+
step 2900000 | 27371s | train_loss=0.0225 | fid=171.3326
|
| 293 |
+
step 2910000 | 27457s | train_loss=0.0216
|
| 294 |
+
step 2920000 | 27546s | train_loss=0.0226
|
| 295 |
+
step 2930000 | 27635s | train_loss=0.0212
|
| 296 |
+
step 2940000 | 27723s | train_loss=0.0261
|
| 297 |
+
step 2950000 | 27812s | train_loss=0.0222
|
| 298 |
+
step 2960000 | 27901s | train_loss=0.0236
|
| 299 |
+
step 2970000 | 27985s | train_loss=0.0223
|
| 300 |
+
step 2980000 | 28067s | train_loss=0.0225
|
| 301 |
+
step 2990000 | 28156s | train_loss=0.0222
|
| 302 |
+
step 3000000 | 28314s | train_loss=0.0235 | fid=163.2940
|
| 303 |
+
step 3010000 | 28393s | train_loss=0.0240
|
| 304 |
+
step 3020000 | 28467s | train_loss=0.0222
|
| 305 |
+
step 3030000 | 28556s | train_loss=0.0217
|
| 306 |
+
step 3040000 | 28644s | train_loss=0.0237
|
| 307 |
+
step 3050000 | 28733s | train_loss=0.0251
|
| 308 |
+
step 3060000 | 28822s | train_loss=0.0242
|
| 309 |
+
step 3070000 | 28912s | train_loss=0.0222
|
| 310 |
+
step 3080000 | 28988s | train_loss=0.0248
|
| 311 |
+
step 3090000 | 29076s | train_loss=0.0225
|
| 312 |
+
step 3100000 | 29234s | train_loss=0.0230 | fid=159.6721
|
| 313 |
+
step 3110000 | 29323s | train_loss=0.0237
|
| 314 |
+
step 3120000 | 29405s | train_loss=0.0219
|
| 315 |
+
step 3130000 | 29485s | train_loss=0.0217
|
| 316 |
+
step 3140000 | 29558s | train_loss=0.0203
|
| 317 |
+
step 3150000 | 29648s | train_loss=0.0209
|
| 318 |
+
step 3160000 | 29737s | train_loss=0.0217
|
| 319 |
+
step 3170000 | 29826s | train_loss=0.0227
|
| 320 |
+
step 3180000 | 29914s | train_loss=0.0271
|
| 321 |
+
step 3190000 | 30001s | train_loss=0.0200
|
| 322 |
+
step 3200000 | 30158s | train_loss=0.0247 | fid=158.3944
|
| 323 |
+
step 3210000 | 30247s | train_loss=0.0206
|
| 324 |
+
step 3220000 | 30336s | train_loss=0.0209
|
| 325 |
+
step 3230000 | 30426s | train_loss=0.0220
|
| 326 |
+
step 3240000 | 30511s | train_loss=0.0221
|
| 327 |
+
step 3250000 | 30598s | train_loss=0.0257
|
| 328 |
+
step 3260000 | 30687s | train_loss=0.0242
|
| 329 |
+
step 3270000 | 30777s | train_loss=0.0218
|
| 330 |
+
step 3280000 | 30866s | train_loss=0.0208
|
| 331 |
+
step 3290000 | 30952s | train_loss=0.0197
|
| 332 |
+
step 3300000 | 31110s | train_loss=0.0210 | fid=145.9514
|
| 333 |
+
step 3310000 | 31199s | train_loss=0.0222
|
| 334 |
+
step 3320000 | 31288s | train_loss=0.0205
|
| 335 |
+
step 3330000 | 31375s | train_loss=0.0218
|
| 336 |
+
step 3340000 | 31465s | train_loss=0.0210
|
| 337 |
+
step 3350000 | 31543s | train_loss=0.0215
|
| 338 |
+
step 3360000 | 31627s | train_loss=0.0205
|
| 339 |
+
step 3370000 | 31707s | train_loss=0.0229
|
| 340 |
+
step 3380000 | 31797s | train_loss=0.0221
|
| 341 |
+
step 3390000 | 31874s | train_loss=0.0217
|
| 342 |
+
step 3400000 | 32031s | train_loss=0.0213 | fid=149.4538
|
| 343 |
+
step 3410000 | 32120s | train_loss=0.0237
|
| 344 |
+
step 3420000 | 32210s | train_loss=0.0224
|
| 345 |
+
step 3430000 | 32297s | train_loss=0.0216
|
| 346 |
+
step 3440000 | 32386s | train_loss=0.0214
|
| 347 |
+
step 3450000 | 32476s | train_loss=0.0240
|
| 348 |
+
step 3460000 | 32562s | train_loss=0.0224
|
| 349 |
+
step 3470000 | 32640s | train_loss=0.0205
|
| 350 |
+
step 3480000 | 32719s | train_loss=0.0204
|
| 351 |
+
step 3490000 | 32800s | train_loss=0.0203
|
| 352 |
+
step 3500000 | 32954s | train_loss=0.0211 | fid=143.2369
|
| 353 |
+
step 3510000 | 33043s | train_loss=0.0205
|
| 354 |
+
step 3520000 | 33132s | train_loss=0.0201
|
| 355 |
+
step 3530000 | 33220s | train_loss=0.0218
|
| 356 |
+
step 3540000 | 33309s | train_loss=0.0234
|
| 357 |
+
step 3550000 | 33399s | train_loss=0.0217
|
| 358 |
+
step 3560000 | 33488s | train_loss=0.0205
|
| 359 |
+
step 3570000 | 33577s | train_loss=0.0212
|
| 360 |
+
step 3580000 | 33657s | train_loss=0.0209
|
| 361 |
+
step 3590000 | 33742s | train_loss=0.0202
|
| 362 |
+
step 3600000 | 33877s | train_loss=0.0206 | fid=151.4144
|
| 363 |
+
step 3610000 | 33966s | train_loss=0.0220
|
| 364 |
+
step 3620000 | 34055s | train_loss=0.0215
|
| 365 |
+
step 3630000 | 34143s | train_loss=0.0214
|
| 366 |
+
step 3640000 | 34232s | train_loss=0.0212
|
| 367 |
+
step 3650000 | 34322s | train_loss=0.0216
|
| 368 |
+
step 3660000 | 34411s | train_loss=0.0219
|
| 369 |
+
step 3670000 | 34500s | train_loss=0.0244
|
| 370 |
+
step 3680000 | 34589s | train_loss=0.0221
|
| 371 |
+
step 3690000 | 34679s | train_loss=0.0215
|
| 372 |
+
step 3700000 | 34808s | train_loss=0.0216 | fid=144.7136
|
| 373 |
+
step 3710000 | 34892s | train_loss=0.0210
|
| 374 |
+
step 3720000 | 34980s | train_loss=0.0214
|
| 375 |
+
step 3730000 | 35069s | train_loss=0.0213
|
| 376 |
+
step 3740000 | 35158s | train_loss=0.0212
|
| 377 |
+
step 3750000 | 35247s | train_loss=0.0213
|
| 378 |
+
step 3760000 | 35336s | train_loss=0.0199
|
| 379 |
+
step 3770000 | 35425s | train_loss=0.0214
|
| 380 |
+
step 3780000 | 35515s | train_loss=0.0214
|
| 381 |
+
step 3790000 | 35604s | train_loss=0.0209
|
| 382 |
+
step 3800000 | 35750s | train_loss=0.0209 | fid=142.0068
|
| 383 |
+
step 3810000 | 35831s | train_loss=0.0221
|
| 384 |
+
step 3820000 | 35908s | train_loss=0.0206
|
| 385 |
+
step 3830000 | 35998s | train_loss=0.0205
|
| 386 |
+
step 3840000 | 36087s | train_loss=0.0203
|
| 387 |
+
step 3850000 | 36176s | train_loss=0.0210
|
| 388 |
+
step 3860000 | 36266s | train_loss=0.0194
|
| 389 |
+
step 3870000 | 36355s | train_loss=0.0225
|
| 390 |
+
step 3880000 | 36444s | train_loss=0.0208
|
| 391 |
+
step 3890000 | 36534s | train_loss=0.0205
|
| 392 |
+
step 3900000 | 36690s | train_loss=0.0217 | fid=142.4868
|
| 393 |
+
step 3910000 | 36767s | train_loss=0.0190
|
| 394 |
+
step 3920000 | 36841s | train_loss=0.0209
|
| 395 |
+
step 3930000 | 36921s | train_loss=0.0196
|
| 396 |
+
step 3940000 | 37005s | train_loss=0.0191
|
| 397 |
+
step 3950000 | 37094s | train_loss=0.0205
|
| 398 |
+
step 3960000 | 37183s | train_loss=0.0205
|
| 399 |
+
step 3970000 | 37272s | train_loss=0.0241
|
| 400 |
+
step 3980000 | 37361s | train_loss=0.0208
|
| 401 |
+
step 3990000 | 37451s | train_loss=0.0194
|
| 402 |
+
step 4000000 | 37608s | train_loss=0.0202 | fid=140.2013
|
| 403 |
+
step 4010000 | 37690s | train_loss=0.0191
|
| 404 |
+
step 4020000 | 37775s | train_loss=0.0256
|
| 405 |
+
step 4030000 | 37862s | train_loss=0.0200
|
| 406 |
+
step 4040000 | 37936s | train_loss=0.0217
|
| 407 |
+
step 4050000 | 38012s | train_loss=0.0201
|
| 408 |
+
step 4060000 | 38098s | train_loss=0.0211
|
| 409 |
+
step 4070000 | 38188s | train_loss=0.0216
|
| 410 |
+
step 4080000 | 38277s | train_loss=0.0197
|
| 411 |
+
step 4090000 | 38366s | train_loss=0.0190
|
| 412 |
+
step 4100000 | 38524s | train_loss=0.0189 | fid=135.9470
|
| 413 |
+
step 4110000 | 38611s | train_loss=0.0230
|
| 414 |
+
step 4120000 | 38692s | train_loss=0.0193
|
| 415 |
+
step 4130000 | 38781s | train_loss=0.0203
|
| 416 |
+
step 4140000 | 38870s | train_loss=0.0195
|
| 417 |
+
step 4150000 | 38954s | train_loss=0.0190
|
| 418 |
+
step 4160000 | 39035s | train_loss=0.0193
|
| 419 |
+
step 4170000 | 39112s | train_loss=0.0223
|
| 420 |
+
step 4180000 | 39201s | train_loss=0.0222
|
| 421 |
+
step 4190000 | 39290s | train_loss=0.0209
|
| 422 |
+
step 4200000 | 39448s | train_loss=0.0180 | fid=134.5767
|
| 423 |
+
step 4210000 | 39535s | train_loss=0.0211
|
| 424 |
+
step 4220000 | 39615s | train_loss=0.0204
|
| 425 |
+
step 4230000 | 39704s | train_loss=0.0184
|
| 426 |
+
step 4240000 | 39793s | train_loss=0.0195
|
| 427 |
+
step 4250000 | 39883s | train_loss=0.0203
|
| 428 |
+
step 4260000 | 39972s | train_loss=0.0212
|
| 429 |
+
step 4270000 | 40049s | train_loss=0.0193
|
| 430 |
+
step 4280000 | 40132s | train_loss=0.0201
|
| 431 |
+
step 4290000 | 40218s | train_loss=0.0200
|
| 432 |
+
step 4300000 | 40376s | train_loss=0.0197 | fid=130.6137
|
| 433 |
+
step 4310000 | 40463s | train_loss=0.0200
|
| 434 |
+
step 4320000 | 40549s | train_loss=0.0191
|
| 435 |
+
step 4330000 | 40630s | train_loss=0.0208
|
| 436 |
+
step 4340000 | 40719s | train_loss=0.0189
|
| 437 |
+
step 4350000 | 40809s | train_loss=0.0193
|
| 438 |
+
step 4360000 | 40898s | train_loss=0.0224
|
| 439 |
+
step 4370000 | 40987s | train_loss=0.0192
|
| 440 |
+
step 4380000 | 41074s | train_loss=0.0209
|
| 441 |
+
step 4390000 | 41152s | train_loss=0.0206
|
| 442 |
+
step 4400000 | 41294s | train_loss=0.0206 | fid=133.0531
|
| 443 |
+
step 4410000 | 41382s | train_loss=0.0216
|
| 444 |
+
step 4420000 | 41471s | train_loss=0.0194
|
| 445 |
+
step 4430000 | 41548s | train_loss=0.0193
|
| 446 |
+
step 4440000 | 41637s | train_loss=0.0201
|
| 447 |
+
step 4450000 | 41726s | train_loss=0.0202
|
| 448 |
+
step 4460000 | 41816s | train_loss=0.0185
|
| 449 |
+
step 4470000 | 41905s | train_loss=0.0197
|
| 450 |
+
step 4480000 | 41994s | train_loss=0.0175
|
| 451 |
+
step 4490000 | 42084s | train_loss=0.0197
|
| 452 |
+
step 4500000 | 42226s | train_loss=0.0181 | fid=135.6460
|
| 453 |
+
step 4510000 | 42296s | train_loss=0.0183
|
| 454 |
+
step 4520000 | 42385s | train_loss=0.0200
|
| 455 |
+
step 4530000 | 42472s | train_loss=0.0224
|
| 456 |
+
step 4540000 | 42548s | train_loss=0.0195
|
| 457 |
+
step 4550000 | 42637s | train_loss=0.0200
|
| 458 |
+
step 4560000 | 42726s | train_loss=0.0204
|
| 459 |
+
step 4570000 | 42815s | train_loss=0.0200
|
| 460 |
+
step 4580000 | 42904s | train_loss=0.0187
|
| 461 |
+
step 4590000 | 42994s | train_loss=0.0209
|
| 462 |
+
step 4600000 | 43148s | train_loss=0.0178 | fid=126.3784
|
| 463 |
+
step 4610000 | 43227s | train_loss=0.0192
|
| 464 |
+
step 4620000 | 43311s | train_loss=0.0191
|
| 465 |
+
step 4630000 | 43392s | train_loss=0.0209
|
| 466 |
+
step 4640000 | 43476s | train_loss=0.0204
|
| 467 |
+
step 4650000 | 43561s | train_loss=0.0196
|
| 468 |
+
step 4660000 | 43650s | train_loss=0.0201
|
| 469 |
+
step 4670000 | 43739s | train_loss=0.0195
|
| 470 |
+
step 4680000 | 43828s | train_loss=0.0181
|
| 471 |
+
step 4690000 | 43918s | train_loss=0.0195
|
| 472 |
+
step 4700000 | 44074s | train_loss=0.0187 | fid=126.5213
|
| 473 |
+
step 4710000 | 44163s | train_loss=0.0205
|
| 474 |
+
step 4720000 | 44252s | train_loss=0.0187
|
| 475 |
+
step 4730000 | 44331s | train_loss=0.0194
|
| 476 |
+
step 4740000 | 44408s | train_loss=0.0188
|
| 477 |
+
step 4750000 | 44483s | train_loss=0.0199
|
| 478 |
+
step 4760000 | 44572s | train_loss=0.0198
|
| 479 |
+
step 4770000 | 44661s | train_loss=0.0187
|
| 480 |
+
step 4780000 | 44750s | train_loss=0.0177
|
| 481 |
+
step 4790000 | 44839s | train_loss=0.0192
|
| 482 |
+
step 4800000 | 44995s | train_loss=0.0194 | fid=132.7748
|
| 483 |
+
step 4810000 | 45085s | train_loss=0.0180
|
| 484 |
+
step 4820000 | 45174s | train_loss=0.0201
|
| 485 |
+
step 4830000 | 45263s | train_loss=0.0186
|
| 486 |
+
step 4840000 | 45345s | train_loss=0.0183
|
| 487 |
+
step 4850000 | 45422s | train_loss=0.0182
|
| 488 |
+
step 4860000 | 45495s | train_loss=0.0193
|
| 489 |
+
step 4870000 | 45585s | train_loss=0.0183
|
| 490 |
+
step 4880000 | 45674s | train_loss=0.0181
|
| 491 |
+
step 4890000 | 45763s | train_loss=0.0187
|
| 492 |
+
step 4900000 | 45905s | train_loss=0.0223 | fid=124.6050
|
| 493 |
+
step 4910000 | 45977s | train_loss=0.0202
|
| 494 |
+
step 4920000 | 46049s | train_loss=0.0185
|
| 495 |
+
step 4930000 | 46120s | train_loss=0.0193
|
| 496 |
+
step 4940000 | 46192s | train_loss=0.0176
|
| 497 |
+
step 4950000 | 46264s | train_loss=0.0190
|
| 498 |
+
step 4960000 | 46307s | train_loss=0.0187
|
| 499 |
+
step 4970000 | 46376s | train_loss=0.0214
|
| 500 |
+
step 4980000 | 46448s | train_loss=0.0187
|
| 501 |
+
step 4990000 | 46519s | train_loss=0.0185
|
| 502 |
+
step 5000000 | 46647s | train_loss=0.0173 | fid=129.0048
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d180/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d180/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d180",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 45,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d180/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d180/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 72s | train_loss=1.0006 | fid=286.1140
|
| 3 |
+
step 10000 | 169s | train_loss=0.7709
|
| 4 |
+
step 20000 | 266s | train_loss=0.6551
|
| 5 |
+
step 30000 | 364s | train_loss=0.5577
|
| 6 |
+
step 40000 | 462s | train_loss=0.4694
|
| 7 |
+
step 50000 | 559s | train_loss=0.3955
|
| 8 |
+
step 60000 | 657s | train_loss=0.3258
|
| 9 |
+
step 70000 | 755s | train_loss=0.2774
|
| 10 |
+
step 80000 | 852s | train_loss=0.2318
|
| 11 |
+
step 90000 | 950s | train_loss=0.2022
|
| 12 |
+
step 100000 | 1116s | train_loss=0.1787 | fid=199.6830
|
| 13 |
+
step 110000 | 1212s | train_loss=0.1542
|
| 14 |
+
step 120000 | 1310s | train_loss=0.1444
|
| 15 |
+
step 130000 | 1407s | train_loss=0.1406
|
| 16 |
+
step 140000 | 1505s | train_loss=0.1296
|
| 17 |
+
step 150000 | 1602s | train_loss=0.1270
|
| 18 |
+
step 160000 | 1699s | train_loss=0.1141
|
| 19 |
+
step 170000 | 1797s | train_loss=0.1066
|
| 20 |
+
step 180000 | 1894s | train_loss=0.1070
|
| 21 |
+
step 190000 | 1984s | train_loss=0.1063
|
| 22 |
+
step 200000 | 2158s | train_loss=0.1114 | fid=260.9392
|
| 23 |
+
step 210000 | 2253s | train_loss=0.1044
|
| 24 |
+
step 220000 | 2351s | train_loss=0.0974
|
| 25 |
+
step 230000 | 2448s | train_loss=0.0880
|
| 26 |
+
step 240000 | 2546s | train_loss=0.0896
|
| 27 |
+
step 250000 | 2643s | train_loss=0.0857
|
| 28 |
+
step 260000 | 2741s | train_loss=0.0808
|
| 29 |
+
step 270000 | 2838s | train_loss=0.0880
|
| 30 |
+
step 280000 | 2931s | train_loss=0.0856
|
| 31 |
+
step 290000 | 3025s | train_loss=0.0791
|
| 32 |
+
step 300000 | 3200s | train_loss=0.0756 | fid=319.0741
|
| 33 |
+
step 310000 | 3295s | train_loss=0.0816
|
| 34 |
+
step 320000 | 3392s | train_loss=0.0718
|
| 35 |
+
step 330000 | 3490s | train_loss=0.0706
|
| 36 |
+
step 340000 | 3587s | train_loss=0.0730
|
| 37 |
+
step 350000 | 3684s | train_loss=0.0678
|
| 38 |
+
step 360000 | 3781s | train_loss=0.0695
|
| 39 |
+
step 370000 | 3876s | train_loss=0.0649
|
| 40 |
+
step 380000 | 3969s | train_loss=0.0642
|
| 41 |
+
step 390000 | 4067s | train_loss=0.0663
|
| 42 |
+
step 400000 | 4241s | train_loss=0.0628 | fid=322.7944
|
| 43 |
+
step 410000 | 4337s | train_loss=0.0617
|
| 44 |
+
step 420000 | 4435s | train_loss=0.0619
|
| 45 |
+
step 430000 | 4532s | train_loss=0.0621
|
| 46 |
+
step 440000 | 4630s | train_loss=0.0594
|
| 47 |
+
step 450000 | 4725s | train_loss=0.0639
|
| 48 |
+
step 460000 | 4821s | train_loss=0.0600
|
| 49 |
+
step 470000 | 4913s | train_loss=0.0569
|
| 50 |
+
step 480000 | 5011s | train_loss=0.0578
|
| 51 |
+
step 490000 | 5108s | train_loss=0.0554
|
| 52 |
+
step 500000 | 5276s | train_loss=0.0570 | fid=315.9428
|
| 53 |
+
step 510000 | 5365s | train_loss=0.0572
|
| 54 |
+
step 520000 | 5457s | train_loss=0.0560
|
| 55 |
+
step 530000 | 5549s | train_loss=0.0553
|
| 56 |
+
step 540000 | 5643s | train_loss=0.0565
|
| 57 |
+
step 550000 | 5738s | train_loss=0.0532
|
| 58 |
+
step 560000 | 5830s | train_loss=0.0622
|
| 59 |
+
step 570000 | 5928s | train_loss=0.0539
|
| 60 |
+
step 580000 | 6025s | train_loss=0.0522
|
| 61 |
+
step 590000 | 6122s | train_loss=0.0531
|
| 62 |
+
step 600000 | 6298s | train_loss=0.0560 | fid=303.4426
|
| 63 |
+
step 610000 | 6393s | train_loss=0.0543
|
| 64 |
+
step 620000 | 6490s | train_loss=0.0558
|
| 65 |
+
step 630000 | 6585s | train_loss=0.0508
|
| 66 |
+
step 640000 | 6683s | train_loss=0.0495
|
| 67 |
+
step 650000 | 6775s | train_loss=0.0541
|
| 68 |
+
step 660000 | 6872s | train_loss=0.0470
|
| 69 |
+
step 670000 | 6970s | train_loss=0.0488
|
| 70 |
+
step 680000 | 7067s | train_loss=0.0535
|
| 71 |
+
step 690000 | 7165s | train_loss=0.0503
|
| 72 |
+
step 700000 | 7341s | train_loss=0.0489 | fid=283.6315
|
| 73 |
+
step 710000 | 7436s | train_loss=0.0485
|
| 74 |
+
step 720000 | 7531s | train_loss=0.0479
|
| 75 |
+
step 730000 | 7628s | train_loss=0.0500
|
| 76 |
+
step 740000 | 7720s | train_loss=0.0471
|
| 77 |
+
step 750000 | 7817s | train_loss=0.0470
|
| 78 |
+
step 760000 | 7914s | train_loss=0.0457
|
| 79 |
+
step 770000 | 8012s | train_loss=0.0434
|
| 80 |
+
step 780000 | 8109s | train_loss=0.0450
|
| 81 |
+
step 790000 | 8206s | train_loss=0.0462
|
| 82 |
+
step 800000 | 8382s | train_loss=0.0436 | fid=279.1933
|
| 83 |
+
step 810000 | 8474s | train_loss=0.0437
|
| 84 |
+
step 820000 | 8571s | train_loss=0.0435
|
| 85 |
+
step 830000 | 8663s | train_loss=0.0423
|
| 86 |
+
step 840000 | 8761s | train_loss=0.0458
|
| 87 |
+
step 850000 | 8858s | train_loss=0.0411
|
| 88 |
+
step 860000 | 8956s | train_loss=0.0449
|
| 89 |
+
step 870000 | 9053s | train_loss=0.0409
|
| 90 |
+
step 880000 | 9151s | train_loss=0.0424
|
| 91 |
+
step 890000 | 9248s | train_loss=0.0440
|
| 92 |
+
step 900000 | 9422s | train_loss=0.0415 | fid=254.8164
|
| 93 |
+
step 910000 | 9517s | train_loss=0.0398
|
| 94 |
+
step 920000 | 9610s | train_loss=0.0443
|
| 95 |
+
step 930000 | 9706s | train_loss=0.0395
|
| 96 |
+
step 940000 | 9804s | train_loss=0.0418
|
| 97 |
+
step 950000 | 9899s | train_loss=0.0421
|
| 98 |
+
step 960000 | 9996s | train_loss=0.0429
|
| 99 |
+
step 970000 | 10093s | train_loss=0.0374
|
| 100 |
+
step 980000 | 10190s | train_loss=0.0411
|
| 101 |
+
step 990000 | 10287s | train_loss=0.0377
|
| 102 |
+
step 1000000 | 10460s | train_loss=0.0410 | fid=238.0060
|
| 103 |
+
step 1010000 | 10551s | train_loss=0.0396
|
| 104 |
+
step 1020000 | 10645s | train_loss=0.0394
|
| 105 |
+
step 1030000 | 10743s | train_loss=0.0405
|
| 106 |
+
step 1040000 | 10840s | train_loss=0.0389
|
| 107 |
+
step 1050000 | 10938s | train_loss=0.0400
|
| 108 |
+
step 1060000 | 11035s | train_loss=0.0394
|
| 109 |
+
step 1070000 | 11132s | train_loss=0.0525
|
| 110 |
+
step 1080000 | 11229s | train_loss=0.0362
|
| 111 |
+
step 1090000 | 11325s | train_loss=0.0387
|
| 112 |
+
step 1100000 | 11500s | train_loss=0.0399 | fid=223.0556
|
| 113 |
+
step 1110000 | 11590s | train_loss=0.0373
|
| 114 |
+
step 1120000 | 11688s | train_loss=0.0391
|
| 115 |
+
step 1130000 | 11785s | train_loss=0.0388
|
| 116 |
+
step 1140000 | 11882s | train_loss=0.0406
|
| 117 |
+
step 1150000 | 11980s | train_loss=0.0381
|
| 118 |
+
step 1160000 | 12077s | train_loss=0.0371
|
| 119 |
+
step 1170000 | 12172s | train_loss=0.0366
|
| 120 |
+
step 1180000 | 12269s | train_loss=0.0363
|
| 121 |
+
step 1190000 | 12367s | train_loss=0.0361
|
| 122 |
+
step 1200000 | 12539s | train_loss=0.0354 | fid=221.5728
|
| 123 |
+
step 1210000 | 12634s | train_loss=0.0344
|
| 124 |
+
step 1220000 | 12731s | train_loss=0.0363
|
| 125 |
+
step 1230000 | 12828s | train_loss=0.0349
|
| 126 |
+
step 1240000 | 12925s | train_loss=0.0377
|
| 127 |
+
step 1250000 | 13022s | train_loss=0.0343
|
| 128 |
+
step 1260000 | 13117s | train_loss=0.0378
|
| 129 |
+
step 1270000 | 13214s | train_loss=0.0344
|
| 130 |
+
step 1280000 | 13312s | train_loss=0.0375
|
| 131 |
+
step 1290000 | 13407s | train_loss=0.0385
|
| 132 |
+
step 1300000 | 13579s | train_loss=0.0326 | fid=196.1262
|
| 133 |
+
step 1310000 | 13675s | train_loss=0.0347
|
| 134 |
+
step 1320000 | 13772s | train_loss=0.0348
|
| 135 |
+
step 1330000 | 13868s | train_loss=0.0332
|
| 136 |
+
step 1340000 | 13965s | train_loss=0.0345
|
| 137 |
+
step 1350000 | 14060s | train_loss=0.0351
|
| 138 |
+
step 1360000 | 14157s | train_loss=0.0321
|
| 139 |
+
step 1370000 | 14255s | train_loss=0.0305
|
| 140 |
+
step 1380000 | 14350s | train_loss=0.0326
|
| 141 |
+
step 1390000 | 14433s | train_loss=0.0340
|
| 142 |
+
step 1400000 | 14642s | train_loss=0.0312 | fid=182.1244
|
| 143 |
+
step 1410000 | 14735s | train_loss=0.0327
|
| 144 |
+
step 1420000 | 14833s | train_loss=0.0325
|
| 145 |
+
step 1430000 | 14930s | train_loss=0.0359
|
| 146 |
+
step 1440000 | 15025s | train_loss=0.0334
|
| 147 |
+
step 1450000 | 15123s | train_loss=0.0330
|
| 148 |
+
step 1460000 | 15220s | train_loss=0.0319
|
| 149 |
+
step 1470000 | 15315s | train_loss=0.0318
|
| 150 |
+
step 1480000 | 15409s | train_loss=0.0309
|
| 151 |
+
step 1490000 | 15507s | train_loss=0.0326
|
| 152 |
+
step 1500000 | 15684s | train_loss=0.0335 | fid=187.5650
|
| 153 |
+
step 1510000 | 15778s | train_loss=0.0335
|
| 154 |
+
step 1520000 | 15873s | train_loss=0.0322
|
| 155 |
+
step 1530000 | 15970s | train_loss=0.0317
|
| 156 |
+
step 1540000 | 16067s | train_loss=0.0305
|
| 157 |
+
step 1550000 | 16164s | train_loss=0.0319
|
| 158 |
+
step 1560000 | 16260s | train_loss=0.0331
|
| 159 |
+
step 1570000 | 16349s | train_loss=0.0308
|
| 160 |
+
step 1580000 | 16442s | train_loss=0.0314
|
| 161 |
+
step 1590000 | 16540s | train_loss=0.0293
|
| 162 |
+
step 1600000 | 16739s | train_loss=0.0306 | fid=177.6150
|
| 163 |
+
step 1610000 | 16820s | train_loss=0.0306
|
| 164 |
+
step 1620000 | 16917s | train_loss=0.0311
|
| 165 |
+
step 1630000 | 17014s | train_loss=0.0320
|
| 166 |
+
step 1640000 | 17112s | train_loss=0.0284
|
| 167 |
+
step 1650000 | 17207s | train_loss=0.0298
|
| 168 |
+
step 1660000 | 17302s | train_loss=0.0315
|
| 169 |
+
step 1670000 | 17385s | train_loss=0.0324
|
| 170 |
+
step 1680000 | 17482s | train_loss=0.0292
|
| 171 |
+
step 1690000 | 17580s | train_loss=0.0327
|
| 172 |
+
step 1700000 | 17795s | train_loss=0.0306 | fid=167.9407
|
| 173 |
+
step 1710000 | 17887s | train_loss=0.0289
|
| 174 |
+
step 1720000 | 17984s | train_loss=0.0315
|
| 175 |
+
step 1730000 | 18079s | train_loss=0.0293
|
| 176 |
+
step 1740000 | 18177s | train_loss=0.0330
|
| 177 |
+
step 1750000 | 18274s | train_loss=0.0293
|
| 178 |
+
step 1760000 | 18368s | train_loss=0.0279
|
| 179 |
+
step 1770000 | 18465s | train_loss=0.0295
|
| 180 |
+
step 1780000 | 18563s | train_loss=0.0318
|
| 181 |
+
step 1790000 | 18658s | train_loss=0.0283
|
| 182 |
+
step 1800000 | 18835s | train_loss=0.0282 | fid=180.0677
|
| 183 |
+
step 1810000 | 18929s | train_loss=0.0316
|
| 184 |
+
step 1820000 | 19025s | train_loss=0.0283
|
| 185 |
+
step 1830000 | 19122s | train_loss=0.0292
|
| 186 |
+
step 1840000 | 19220s | train_loss=0.0279
|
| 187 |
+
step 1850000 | 19313s | train_loss=0.0256
|
| 188 |
+
step 1860000 | 19411s | train_loss=0.0273
|
| 189 |
+
step 1870000 | 19508s | train_loss=0.0301
|
| 190 |
+
step 1880000 | 19603s | train_loss=0.0281
|
| 191 |
+
step 1890000 | 19700s | train_loss=0.0294
|
| 192 |
+
step 1900000 | 19916s | train_loss=0.0276 | fid=169.5027
|
| 193 |
+
step 1910000 | 20005s | train_loss=0.0299
|
| 194 |
+
step 1920000 | 20102s | train_loss=0.0289
|
| 195 |
+
step 1930000 | 20195s | train_loss=0.0281
|
| 196 |
+
step 1940000 | 20284s | train_loss=0.0256
|
| 197 |
+
step 1950000 | 20382s | train_loss=0.0271
|
| 198 |
+
step 1960000 | 20477s | train_loss=0.0275
|
| 199 |
+
step 1970000 | 20574s | train_loss=0.0272
|
| 200 |
+
step 1980000 | 20671s | train_loss=0.0268
|
| 201 |
+
step 1990000 | 20768s | train_loss=0.0306
|
| 202 |
+
step 2000000 | 20978s | train_loss=0.0293 | fid=170.8974
|
| 203 |
+
step 2010000 | 21070s | train_loss=0.0269
|
| 204 |
+
step 2020000 | 21164s | train_loss=0.0256
|
| 205 |
+
step 2030000 | 21256s | train_loss=0.0258
|
| 206 |
+
step 2040000 | 21352s | train_loss=0.0262
|
| 207 |
+
step 2050000 | 21448s | train_loss=0.0291
|
| 208 |
+
step 2060000 | 21545s | train_loss=0.0291
|
| 209 |
+
step 2070000 | 21642s | train_loss=0.0319
|
| 210 |
+
step 2080000 | 21739s | train_loss=0.0271
|
| 211 |
+
step 2090000 | 21835s | train_loss=0.0262
|
| 212 |
+
step 2100000 | 22047s | train_loss=0.0279 | fid=168.6363
|
| 213 |
+
step 2110000 | 22131s | train_loss=0.0289
|
| 214 |
+
step 2120000 | 22219s | train_loss=0.0273
|
| 215 |
+
step 2130000 | 22315s | train_loss=0.0259
|
| 216 |
+
step 2140000 | 22412s | train_loss=0.0276
|
| 217 |
+
step 2150000 | 22509s | train_loss=0.0283
|
| 218 |
+
step 2160000 | 22606s | train_loss=0.0263
|
| 219 |
+
step 2170000 | 22704s | train_loss=0.0289
|
| 220 |
+
step 2180000 | 22799s | train_loss=0.0274
|
| 221 |
+
step 2190000 | 22896s | train_loss=0.0276
|
| 222 |
+
step 2200000 | 23090s | train_loss=0.0265 | fid=170.3191
|
| 223 |
+
step 2210000 | 23161s | train_loss=0.0255
|
| 224 |
+
step 2220000 | 23258s | train_loss=0.0258
|
| 225 |
+
step 2230000 | 23355s | train_loss=0.0249
|
| 226 |
+
step 2240000 | 23453s | train_loss=0.0233
|
| 227 |
+
step 2250000 | 23550s | train_loss=0.0252
|
| 228 |
+
step 2260000 | 23647s | train_loss=0.0262
|
| 229 |
+
step 2270000 | 23743s | train_loss=0.0267
|
| 230 |
+
step 2280000 | 23840s | train_loss=0.0251
|
| 231 |
+
step 2290000 | 23938s | train_loss=0.0294
|
| 232 |
+
step 2300000 | 24126s | train_loss=0.0246 | fid=172.5928
|
| 233 |
+
step 2310000 | 24209s | train_loss=0.0271
|
| 234 |
+
step 2320000 | 24306s | train_loss=0.0253
|
| 235 |
+
step 2330000 | 24403s | train_loss=0.0289
|
| 236 |
+
step 2340000 | 24500s | train_loss=0.0250
|
| 237 |
+
step 2350000 | 24596s | train_loss=0.0250
|
| 238 |
+
step 2360000 | 24692s | train_loss=0.0271
|
| 239 |
+
step 2370000 | 24790s | train_loss=0.0261
|
| 240 |
+
step 2380000 | 24887s | train_loss=0.0267
|
| 241 |
+
step 2390000 | 24984s | train_loss=0.0242
|
| 242 |
+
step 2400000 | 25189s | train_loss=0.0259 | fid=164.3911
|
| 243 |
+
step 2410000 | 25275s | train_loss=0.0253
|
| 244 |
+
step 2420000 | 25373s | train_loss=0.0239
|
| 245 |
+
step 2430000 | 25470s | train_loss=0.0258
|
| 246 |
+
step 2440000 | 25565s | train_loss=0.0238
|
| 247 |
+
step 2450000 | 25662s | train_loss=0.0251
|
| 248 |
+
step 2460000 | 25760s | train_loss=0.0295
|
| 249 |
+
step 2470000 | 25857s | train_loss=0.0245
|
| 250 |
+
step 2480000 | 25953s | train_loss=0.0243
|
| 251 |
+
step 2490000 | 26050s | train_loss=0.0253
|
| 252 |
+
step 2500000 | 26250s | train_loss=0.0261 | fid=150.0684
|
| 253 |
+
step 2510000 | 26329s | train_loss=0.0248
|
| 254 |
+
step 2520000 | 26426s | train_loss=0.0239
|
| 255 |
+
step 2530000 | 26521s | train_loss=0.0253
|
| 256 |
+
step 2540000 | 26618s | train_loss=0.0262
|
| 257 |
+
step 2550000 | 26716s | train_loss=0.0269
|
| 258 |
+
step 2560000 | 26811s | train_loss=0.0246
|
| 259 |
+
step 2570000 | 26909s | train_loss=0.0259
|
| 260 |
+
step 2580000 | 27005s | train_loss=0.0249
|
| 261 |
+
step 2590000 | 27099s | train_loss=0.0292
|
| 262 |
+
step 2600000 | 27320s | train_loss=0.0254 | fid=160.2681
|
| 263 |
+
step 2610000 | 27403s | train_loss=0.0241
|
| 264 |
+
step 2620000 | 27500s | train_loss=0.0258
|
| 265 |
+
step 2630000 | 27598s | train_loss=0.0249
|
| 266 |
+
step 2640000 | 27695s | train_loss=0.0261
|
| 267 |
+
step 2650000 | 27791s | train_loss=0.0250
|
| 268 |
+
step 2660000 | 27888s | train_loss=0.0254
|
| 269 |
+
step 2670000 | 27980s | train_loss=0.0240
|
| 270 |
+
step 2680000 | 28070s | train_loss=0.0230
|
| 271 |
+
step 2690000 | 28167s | train_loss=0.0229
|
| 272 |
+
step 2700000 | 28369s | train_loss=0.0233 | fid=153.2712
|
| 273 |
+
step 2710000 | 28452s | train_loss=0.0268
|
| 274 |
+
step 2720000 | 28547s | train_loss=0.0233
|
| 275 |
+
step 2730000 | 28643s | train_loss=0.0277
|
| 276 |
+
step 2740000 | 28740s | train_loss=0.0240
|
| 277 |
+
step 2750000 | 28837s | train_loss=0.0228
|
| 278 |
+
step 2760000 | 28934s | train_loss=0.0249
|
| 279 |
+
step 2770000 | 29017s | train_loss=0.0235
|
| 280 |
+
step 2780000 | 29114s | train_loss=0.0233
|
| 281 |
+
step 2790000 | 29211s | train_loss=0.0253
|
| 282 |
+
step 2800000 | 29431s | train_loss=0.0234 | fid=151.5065
|
| 283 |
+
step 2810000 | 29514s | train_loss=0.0255
|
| 284 |
+
step 2820000 | 29607s | train_loss=0.0246
|
| 285 |
+
step 2830000 | 29705s | train_loss=0.0253
|
| 286 |
+
step 2840000 | 29803s | train_loss=0.0228
|
| 287 |
+
step 2850000 | 29900s | train_loss=0.0230
|
| 288 |
+
step 2860000 | 29994s | train_loss=0.0254
|
| 289 |
+
step 2870000 | 30091s | train_loss=0.0241
|
| 290 |
+
step 2880000 | 30186s | train_loss=0.0295
|
| 291 |
+
step 2890000 | 30284s | train_loss=0.0224
|
| 292 |
+
step 2900000 | 30459s | train_loss=0.0216 | fid=154.2750
|
| 293 |
+
step 2910000 | 30557s | train_loss=0.0265
|
| 294 |
+
step 2920000 | 30652s | train_loss=0.0228
|
| 295 |
+
step 2930000 | 30749s | train_loss=0.0246
|
| 296 |
+
step 2940000 | 30846s | train_loss=0.0224
|
| 297 |
+
step 2950000 | 30940s | train_loss=0.0254
|
| 298 |
+
step 2960000 | 31037s | train_loss=0.0225
|
| 299 |
+
step 2970000 | 31132s | train_loss=0.0223
|
| 300 |
+
step 2980000 | 31230s | train_loss=0.0244
|
| 301 |
+
step 2990000 | 31327s | train_loss=0.0246
|
| 302 |
+
step 3000000 | 31536s | train_loss=0.0266 | fid=152.0357
|
| 303 |
+
step 3010000 | 31628s | train_loss=0.0242
|
| 304 |
+
step 3020000 | 31717s | train_loss=0.0221
|
| 305 |
+
step 3030000 | 31812s | train_loss=0.0226
|
| 306 |
+
step 3040000 | 31899s | train_loss=0.0233
|
| 307 |
+
step 3050000 | 31996s | train_loss=0.0235
|
| 308 |
+
step 3060000 | 32092s | train_loss=0.0224
|
| 309 |
+
step 3070000 | 32189s | train_loss=0.0227
|
| 310 |
+
step 3080000 | 32284s | train_loss=0.0242
|
| 311 |
+
step 3090000 | 32382s | train_loss=0.0203
|
| 312 |
+
step 3100000 | 32605s | train_loss=0.0242 | fid=158.0143
|
| 313 |
+
step 3110000 | 32698s | train_loss=0.0223
|
| 314 |
+
step 3120000 | 32785s | train_loss=0.0261
|
| 315 |
+
step 3130000 | 32874s | train_loss=0.0236
|
| 316 |
+
step 3140000 | 32969s | train_loss=0.0225
|
| 317 |
+
step 3150000 | 33066s | train_loss=0.0236
|
| 318 |
+
step 3160000 | 33163s | train_loss=0.0262
|
| 319 |
+
step 3170000 | 33258s | train_loss=0.0216
|
| 320 |
+
step 3180000 | 33356s | train_loss=0.0234
|
| 321 |
+
step 3190000 | 33453s | train_loss=0.0246
|
| 322 |
+
step 3200000 | 33673s | train_loss=0.0218 | fid=148.4415
|
| 323 |
+
step 3210000 | 33760s | train_loss=0.0213
|
| 324 |
+
step 3220000 | 33843s | train_loss=0.0234
|
| 325 |
+
step 3230000 | 33938s | train_loss=0.0224
|
| 326 |
+
step 3240000 | 34035s | train_loss=0.0236
|
| 327 |
+
step 3250000 | 34130s | train_loss=0.0223
|
| 328 |
+
step 3260000 | 34228s | train_loss=0.0204
|
| 329 |
+
step 3270000 | 34325s | train_loss=0.0236
|
| 330 |
+
step 3280000 | 34423s | train_loss=0.0216
|
| 331 |
+
step 3290000 | 34520s | train_loss=0.0225
|
| 332 |
+
step 3300000 | 34733s | train_loss=0.0237 | fid=154.3641
|
| 333 |
+
step 3310000 | 34808s | train_loss=0.0217
|
| 334 |
+
step 3320000 | 34900s | train_loss=0.0218
|
| 335 |
+
step 3330000 | 34995s | train_loss=0.0217
|
| 336 |
+
step 3340000 | 35092s | train_loss=0.0248
|
| 337 |
+
step 3350000 | 35189s | train_loss=0.0204
|
| 338 |
+
step 3360000 | 35287s | train_loss=0.0252
|
| 339 |
+
step 3370000 | 35383s | train_loss=0.0225
|
| 340 |
+
step 3380000 | 35480s | train_loss=0.0223
|
| 341 |
+
step 3390000 | 35578s | train_loss=0.0213
|
| 342 |
+
step 3400000 | 35784s | train_loss=0.0217 | fid=141.4295
|
| 343 |
+
step 3410000 | 35872s | train_loss=0.0238
|
| 344 |
+
step 3420000 | 35965s | train_loss=0.0211
|
| 345 |
+
step 3430000 | 36063s | train_loss=0.0239
|
| 346 |
+
step 3440000 | 36161s | train_loss=0.0200
|
| 347 |
+
step 3450000 | 36258s | train_loss=0.0226
|
| 348 |
+
step 3460000 | 36356s | train_loss=0.0224
|
| 349 |
+
step 3470000 | 36454s | train_loss=0.0225
|
| 350 |
+
step 3480000 | 36551s | train_loss=0.0218
|
| 351 |
+
step 3490000 | 36647s | train_loss=0.0238
|
| 352 |
+
step 3500000 | 36845s | train_loss=0.0220 | fid=150.8439
|
| 353 |
+
step 3510000 | 36930s | train_loss=0.0213
|
| 354 |
+
step 3520000 | 37024s | train_loss=0.0227
|
| 355 |
+
step 3530000 | 37121s | train_loss=0.0213
|
| 356 |
+
step 3540000 | 37219s | train_loss=0.0227
|
| 357 |
+
step 3550000 | 37316s | train_loss=0.0226
|
| 358 |
+
step 3560000 | 37413s | train_loss=0.0208
|
| 359 |
+
step 3570000 | 37511s | train_loss=0.0229
|
| 360 |
+
step 3580000 | 37606s | train_loss=0.0199
|
| 361 |
+
step 3590000 | 37693s | train_loss=0.0236
|
| 362 |
+
step 3600000 | 37922s | train_loss=0.0212 | fid=144.3535
|
| 363 |
+
step 3610000 | 38007s | train_loss=0.0220
|
| 364 |
+
step 3620000 | 38100s | train_loss=0.0206
|
| 365 |
+
step 3630000 | 38198s | train_loss=0.0218
|
| 366 |
+
step 3640000 | 38295s | train_loss=0.0224
|
| 367 |
+
step 3650000 | 38392s | train_loss=0.0229
|
| 368 |
+
step 3660000 | 38490s | train_loss=0.0219
|
| 369 |
+
step 3670000 | 38585s | train_loss=0.0215
|
| 370 |
+
step 3680000 | 38671s | train_loss=0.0224
|
| 371 |
+
step 3690000 | 38768s | train_loss=0.0213
|
| 372 |
+
step 3700000 | 38989s | train_loss=0.0218 | fid=152.7365
|
| 373 |
+
step 3710000 | 39077s | train_loss=0.0223
|
| 374 |
+
step 3720000 | 39170s | train_loss=0.0217
|
| 375 |
+
step 3730000 | 39268s | train_loss=0.0219
|
| 376 |
+
step 3740000 | 39365s | train_loss=0.0204
|
| 377 |
+
step 3750000 | 39460s | train_loss=0.0226
|
| 378 |
+
step 3760000 | 39555s | train_loss=0.0208
|
| 379 |
+
step 3770000 | 39642s | train_loss=0.0203
|
| 380 |
+
step 3780000 | 39739s | train_loss=0.0214
|
| 381 |
+
step 3790000 | 39836s | train_loss=0.0212
|
| 382 |
+
step 3800000 | 40052s | train_loss=0.0224 | fid=152.1230
|
| 383 |
+
step 3810000 | 40141s | train_loss=0.0205
|
| 384 |
+
step 3820000 | 40236s | train_loss=0.0219
|
| 385 |
+
step 3830000 | 40334s | train_loss=0.0245
|
| 386 |
+
step 3840000 | 40427s | train_loss=0.0191
|
| 387 |
+
step 3850000 | 40524s | train_loss=0.0254
|
| 388 |
+
step 3860000 | 40608s | train_loss=0.0213
|
| 389 |
+
step 3870000 | 40705s | train_loss=0.0200
|
| 390 |
+
step 3880000 | 40803s | train_loss=0.0205
|
| 391 |
+
step 3890000 | 40900s | train_loss=0.0211
|
| 392 |
+
step 3900000 | 41122s | train_loss=0.0206 | fid=140.6504
|
| 393 |
+
step 3910000 | 41207s | train_loss=0.0209
|
| 394 |
+
step 3920000 | 41299s | train_loss=0.0205
|
| 395 |
+
step 3930000 | 41395s | train_loss=0.0205
|
| 396 |
+
step 3940000 | 41491s | train_loss=0.0193
|
| 397 |
+
step 3950000 | 41575s | train_loss=0.0211
|
| 398 |
+
step 3960000 | 41672s | train_loss=0.0193
|
| 399 |
+
step 3970000 | 41769s | train_loss=0.0198
|
| 400 |
+
step 3980000 | 41866s | train_loss=0.0258
|
| 401 |
+
step 3990000 | 41963s | train_loss=0.0182
|
| 402 |
+
step 4000000 | 42189s | train_loss=0.0204 | fid=146.6873
|
| 403 |
+
step 4010000 | 42271s | train_loss=0.0207
|
| 404 |
+
step 4020000 | 42361s | train_loss=0.0215
|
| 405 |
+
step 4030000 | 42457s | train_loss=0.0198
|
| 406 |
+
step 4040000 | 42538s | train_loss=0.0196
|
| 407 |
+
step 4050000 | 42635s | train_loss=0.0209
|
| 408 |
+
step 4060000 | 42732s | train_loss=0.0205
|
| 409 |
+
step 4070000 | 42829s | train_loss=0.0201
|
| 410 |
+
step 4080000 | 42926s | train_loss=0.0218
|
| 411 |
+
step 4090000 | 43023s | train_loss=0.0209
|
| 412 |
+
step 4100000 | 43248s | train_loss=0.0190 | fid=155.7251
|
| 413 |
+
step 4110000 | 43337s | train_loss=0.0223
|
| 414 |
+
step 4120000 | 43434s | train_loss=0.0241
|
| 415 |
+
step 4130000 | 43520s | train_loss=0.0197
|
| 416 |
+
step 4140000 | 43617s | train_loss=0.0212
|
| 417 |
+
step 4150000 | 43714s | train_loss=0.0193
|
| 418 |
+
step 4160000 | 43812s | train_loss=0.0202
|
| 419 |
+
step 4170000 | 43909s | train_loss=0.0214
|
| 420 |
+
step 4180000 | 44006s | train_loss=0.0208
|
| 421 |
+
step 4190000 | 44100s | train_loss=0.0206
|
| 422 |
+
step 4200000 | 44308s | train_loss=0.0191 | fid=137.7774
|
| 423 |
+
step 4210000 | 44396s | train_loss=0.0198
|
| 424 |
+
step 4220000 | 44475s | train_loss=0.0199
|
| 425 |
+
step 4230000 | 44571s | train_loss=0.0194
|
| 426 |
+
step 4240000 | 44668s | train_loss=0.0187
|
| 427 |
+
step 4250000 | 44765s | train_loss=0.0208
|
| 428 |
+
step 4260000 | 44862s | train_loss=0.0194
|
| 429 |
+
step 4270000 | 44958s | train_loss=0.0207
|
| 430 |
+
step 4280000 | 45054s | train_loss=0.0203
|
| 431 |
+
step 4290000 | 45151s | train_loss=0.0223
|
| 432 |
+
step 4300000 | 45372s | train_loss=0.0202 | fid=137.1396
|
| 433 |
+
step 4310000 | 45450s | train_loss=0.0196
|
| 434 |
+
step 4320000 | 45543s | train_loss=0.0189
|
| 435 |
+
step 4330000 | 45641s | train_loss=0.0189
|
| 436 |
+
step 4340000 | 45738s | train_loss=0.0188
|
| 437 |
+
step 4350000 | 45835s | train_loss=0.0226
|
| 438 |
+
step 4360000 | 45912s | train_loss=0.0207
|
| 439 |
+
step 4370000 | 45990s | train_loss=0.0194
|
| 440 |
+
step 4380000 | 46068s | train_loss=0.0199
|
| 441 |
+
step 4390000 | 46146s | train_loss=0.0203
|
| 442 |
+
step 4400000 | 46310s | train_loss=0.0214 | fid=146.8397
|
| 443 |
+
step 4410000 | 46389s | train_loss=0.0201
|
| 444 |
+
step 4420000 | 46468s | train_loss=0.0195
|
| 445 |
+
step 4430000 | 46546s | train_loss=0.0223
|
| 446 |
+
step 4440000 | 46624s | train_loss=0.0224
|
| 447 |
+
step 4450000 | 46688s | train_loss=0.0195
|
| 448 |
+
step 4460000 | 46747s | train_loss=0.0241
|
| 449 |
+
step 4470000 | 46807s | train_loss=0.0204
|
| 450 |
+
step 4480000 | 46866s | train_loss=0.0196
|
| 451 |
+
step 4490000 | 46926s | train_loss=0.0192
|
| 452 |
+
step 4500000 | 47034s | train_loss=0.0188 | fid=133.5049
|
| 453 |
+
step 4510000 | 47090s | train_loss=0.0187
|
| 454 |
+
step 4520000 | 47149s | train_loss=0.0196
|
| 455 |
+
step 4530000 | 47209s | train_loss=0.0185
|
| 456 |
+
step 4540000 | 47268s | train_loss=0.0207
|
| 457 |
+
step 4550000 | 47328s | train_loss=0.0189
|
| 458 |
+
step 4560000 | 47387s | train_loss=0.0203
|
| 459 |
+
step 4570000 | 47447s | train_loss=0.0207
|
| 460 |
+
step 4580000 | 47506s | train_loss=0.0201
|
| 461 |
+
step 4590000 | 47561s | train_loss=0.0209
|
| 462 |
+
step 4600000 | 47638s | train_loss=0.0181 | fid=138.0517
|
| 463 |
+
step 4610000 | 47675s | train_loss=0.0211
|
| 464 |
+
step 4620000 | 47716s | train_loss=0.0185
|
| 465 |
+
step 4630000 | 47757s | train_loss=0.0195
|
| 466 |
+
step 4640000 | 47798s | train_loss=0.0184
|
| 467 |
+
step 4650000 | 47838s | train_loss=0.0184
|
| 468 |
+
step 4660000 | 47879s | train_loss=0.0200
|
| 469 |
+
step 4670000 | 47920s | train_loss=0.0201
|
| 470 |
+
step 4680000 | 47961s | train_loss=0.0192
|
| 471 |
+
step 4690000 | 48002s | train_loss=0.0211
|
| 472 |
+
step 4700000 | 48079s | train_loss=0.0197 | fid=143.3286
|
| 473 |
+
step 4710000 | 48116s | train_loss=0.0193
|
| 474 |
+
step 4720000 | 48157s | train_loss=0.0191
|
| 475 |
+
step 4730000 | 48198s | train_loss=0.0196
|
| 476 |
+
step 4740000 | 48238s | train_loss=0.0185
|
| 477 |
+
step 4750000 | 48279s | train_loss=0.0175
|
| 478 |
+
step 4760000 | 48320s | train_loss=0.0232
|
| 479 |
+
step 4770000 | 48361s | train_loss=0.0189
|
| 480 |
+
step 4780000 | 48402s | train_loss=0.0180
|
| 481 |
+
step 4790000 | 48442s | train_loss=0.0186
|
| 482 |
+
step 4800000 | 48519s | train_loss=0.0184 | fid=136.8204
|
| 483 |
+
step 4810000 | 48556s | train_loss=0.0205
|
| 484 |
+
step 4820000 | 48596s | train_loss=0.0211
|
| 485 |
+
step 4830000 | 48637s | train_loss=0.0198
|
| 486 |
+
step 4840000 | 48678s | train_loss=0.0201
|
| 487 |
+
step 4850000 | 48719s | train_loss=0.0179
|
| 488 |
+
step 4860000 | 48760s | train_loss=0.0202
|
| 489 |
+
step 4870000 | 48800s | train_loss=0.0196
|
| 490 |
+
step 4880000 | 48841s | train_loss=0.0191
|
| 491 |
+
step 4890000 | 48882s | train_loss=0.0200
|
| 492 |
+
step 4900000 | 48958s | train_loss=0.0195 | fid=138.6817
|
| 493 |
+
step 4910000 | 48999s | train_loss=0.0190
|
| 494 |
+
step 4920000 | 49040s | train_loss=0.0196
|
| 495 |
+
step 4930000 | 49081s | train_loss=0.0202
|
| 496 |
+
step 4940000 | 49121s | train_loss=0.0230
|
| 497 |
+
step 4950000 | 49162s | train_loss=0.0194
|
| 498 |
+
step 4960000 | 49203s | train_loss=0.0213
|
| 499 |
+
step 4970000 | 49244s | train_loss=0.0202
|
| 500 |
+
step 4980000 | 49284s | train_loss=0.0205
|
| 501 |
+
step 4990000 | 49325s | train_loss=0.0185
|
| 502 |
+
step 5000000 | 49384s | train_loss=0.0187 | fid=126.1331
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d20/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d20/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d20",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 45,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d20/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d20/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 71s | train_loss=1.0031 | fid=182.4495
|
| 3 |
+
step 10000 | 162s | train_loss=0.9154
|
| 4 |
+
step 20000 | 254s | train_loss=0.9104
|
| 5 |
+
step 30000 | 345s | train_loss=0.9057
|
| 6 |
+
step 40000 | 436s | train_loss=0.9005
|
| 7 |
+
step 50000 | 528s | train_loss=0.8941
|
| 8 |
+
step 60000 | 619s | train_loss=0.8837
|
| 9 |
+
step 70000 | 710s | train_loss=0.8635
|
| 10 |
+
step 80000 | 802s | train_loss=0.8259
|
| 11 |
+
step 90000 | 893s | train_loss=0.7644
|
| 12 |
+
step 100000 | 1050s | train_loss=0.6684 | fid=162.8887
|
| 13 |
+
step 110000 | 1141s | train_loss=0.5526
|
| 14 |
+
step 120000 | 1232s | train_loss=0.4405
|
| 15 |
+
step 130000 | 1323s | train_loss=0.3508
|
| 16 |
+
step 140000 | 1415s | train_loss=0.2829
|
| 17 |
+
step 150000 | 1506s | train_loss=0.2373
|
| 18 |
+
step 160000 | 1597s | train_loss=0.2072
|
| 19 |
+
step 170000 | 1689s | train_loss=0.1891
|
| 20 |
+
step 180000 | 1780s | train_loss=0.1719
|
| 21 |
+
step 190000 | 1871s | train_loss=0.1622
|
| 22 |
+
step 200000 | 2029s | train_loss=0.1514 | fid=164.4031
|
| 23 |
+
step 210000 | 2119s | train_loss=0.1446
|
| 24 |
+
step 220000 | 2210s | train_loss=0.1372
|
| 25 |
+
step 230000 | 2301s | train_loss=0.1293
|
| 26 |
+
step 240000 | 2393s | train_loss=0.1223
|
| 27 |
+
step 250000 | 2484s | train_loss=0.1220
|
| 28 |
+
step 260000 | 2575s | train_loss=0.1131
|
| 29 |
+
step 270000 | 2667s | train_loss=0.1097
|
| 30 |
+
step 280000 | 2758s | train_loss=0.1052
|
| 31 |
+
step 290000 | 2844s | train_loss=0.1036
|
| 32 |
+
step 300000 | 3007s | train_loss=0.0995 | fid=163.2920
|
| 33 |
+
step 310000 | 3096s | train_loss=0.0945
|
| 34 |
+
step 320000 | 3188s | train_loss=0.0930
|
| 35 |
+
step 330000 | 3279s | train_loss=0.0899
|
| 36 |
+
step 340000 | 3370s | train_loss=0.0883
|
| 37 |
+
step 350000 | 3462s | train_loss=0.0847
|
| 38 |
+
step 360000 | 3553s | train_loss=0.0844
|
| 39 |
+
step 370000 | 3645s | train_loss=0.0814
|
| 40 |
+
step 380000 | 3736s | train_loss=0.0797
|
| 41 |
+
step 390000 | 3820s | train_loss=0.0775
|
| 42 |
+
step 400000 | 3985s | train_loss=0.0755 | fid=159.6517
|
| 43 |
+
step 410000 | 4075s | train_loss=0.0747
|
| 44 |
+
step 420000 | 4166s | train_loss=0.0727
|
| 45 |
+
step 430000 | 4257s | train_loss=0.0709
|
| 46 |
+
step 440000 | 4348s | train_loss=0.0705
|
| 47 |
+
step 450000 | 4440s | train_loss=0.0689
|
| 48 |
+
step 460000 | 4531s | train_loss=0.0676
|
| 49 |
+
step 470000 | 4623s | train_loss=0.0667
|
| 50 |
+
step 480000 | 4708s | train_loss=0.0649
|
| 51 |
+
step 490000 | 4797s | train_loss=0.0626
|
| 52 |
+
step 500000 | 4962s | train_loss=0.0632 | fid=159.7633
|
| 53 |
+
step 510000 | 5051s | train_loss=0.0618
|
| 54 |
+
step 520000 | 5142s | train_loss=0.0608
|
| 55 |
+
step 530000 | 5229s | train_loss=0.0591
|
| 56 |
+
step 540000 | 5315s | train_loss=0.0584
|
| 57 |
+
step 550000 | 5403s | train_loss=0.0579
|
| 58 |
+
step 560000 | 5492s | train_loss=0.0569
|
| 59 |
+
step 570000 | 5578s | train_loss=0.0558
|
| 60 |
+
step 580000 | 5661s | train_loss=0.0552
|
| 61 |
+
step 590000 | 5752s | train_loss=0.0541
|
| 62 |
+
step 600000 | 5917s | train_loss=0.0533 | fid=157.1289
|
| 63 |
+
step 610000 | 6007s | train_loss=0.0550
|
| 64 |
+
step 620000 | 6098s | train_loss=0.0530
|
| 65 |
+
step 630000 | 6189s | train_loss=0.0511
|
| 66 |
+
step 640000 | 6281s | train_loss=0.0512
|
| 67 |
+
step 650000 | 6371s | train_loss=0.0515
|
| 68 |
+
step 660000 | 6463s | train_loss=0.0499
|
| 69 |
+
step 670000 | 6548s | train_loss=0.0488
|
| 70 |
+
step 680000 | 6637s | train_loss=0.0488
|
| 71 |
+
step 690000 | 6728s | train_loss=0.0495
|
| 72 |
+
step 700000 | 6893s | train_loss=0.0472 | fid=157.8768
|
| 73 |
+
step 710000 | 6983s | train_loss=0.0475
|
| 74 |
+
step 720000 | 7074s | train_loss=0.0466
|
| 75 |
+
step 730000 | 7166s | train_loss=0.0461
|
| 76 |
+
step 740000 | 7257s | train_loss=0.0461
|
| 77 |
+
step 750000 | 7348s | train_loss=0.0446
|
| 78 |
+
step 760000 | 7436s | train_loss=0.0451
|
| 79 |
+
step 770000 | 7524s | train_loss=0.0447
|
| 80 |
+
step 780000 | 7615s | train_loss=0.0441
|
| 81 |
+
step 790000 | 7706s | train_loss=0.0437
|
| 82 |
+
step 800000 | 7871s | train_loss=0.0426 | fid=157.0994
|
| 83 |
+
step 810000 | 7961s | train_loss=0.0429
|
| 84 |
+
step 820000 | 8052s | train_loss=0.0426
|
| 85 |
+
step 830000 | 8143s | train_loss=0.0415
|
| 86 |
+
step 840000 | 8235s | train_loss=0.0408
|
| 87 |
+
step 850000 | 8326s | train_loss=0.0416
|
| 88 |
+
step 860000 | 8412s | train_loss=0.0406
|
| 89 |
+
step 870000 | 8501s | train_loss=0.0397
|
| 90 |
+
step 880000 | 8592s | train_loss=0.0405
|
| 91 |
+
step 890000 | 8683s | train_loss=0.0397
|
| 92 |
+
step 900000 | 8847s | train_loss=0.0395 | fid=157.5104
|
| 93 |
+
step 910000 | 8937s | train_loss=0.0383
|
| 94 |
+
step 920000 | 9029s | train_loss=0.0386
|
| 95 |
+
step 930000 | 9120s | train_loss=0.0384
|
| 96 |
+
step 940000 | 9212s | train_loss=0.0382
|
| 97 |
+
step 950000 | 9297s | train_loss=0.0382
|
| 98 |
+
step 960000 | 9386s | train_loss=0.0371
|
| 99 |
+
step 970000 | 9477s | train_loss=0.0373
|
| 100 |
+
step 980000 | 9568s | train_loss=0.0362
|
| 101 |
+
step 990000 | 9659s | train_loss=0.0374
|
| 102 |
+
step 1000000 | 9824s | train_loss=0.0361 | fid=155.8247
|
| 103 |
+
step 1010000 | 9914s | train_loss=0.0355
|
| 104 |
+
step 1020000 | 10005s | train_loss=0.0358
|
| 105 |
+
step 1030000 | 10096s | train_loss=0.0355
|
| 106 |
+
step 1040000 | 10186s | train_loss=0.0352
|
| 107 |
+
step 1050000 | 10274s | train_loss=0.0353
|
| 108 |
+
step 1060000 | 10362s | train_loss=0.0356
|
| 109 |
+
step 1070000 | 10454s | train_loss=0.0348
|
| 110 |
+
step 1080000 | 10544s | train_loss=0.0347
|
| 111 |
+
step 1090000 | 10635s | train_loss=0.0345
|
| 112 |
+
step 1100000 | 10800s | train_loss=0.0342 | fid=155.7263
|
| 113 |
+
step 1110000 | 10890s | train_loss=0.0332
|
| 114 |
+
step 1120000 | 10981s | train_loss=0.0335
|
| 115 |
+
step 1130000 | 11072s | train_loss=0.0342
|
| 116 |
+
step 1140000 | 11159s | train_loss=0.0328
|
| 117 |
+
step 1150000 | 11248s | train_loss=0.0331
|
| 118 |
+
step 1160000 | 11339s | train_loss=0.0331
|
| 119 |
+
step 1170000 | 11431s | train_loss=0.0324
|
| 120 |
+
step 1180000 | 11522s | train_loss=0.0325
|
| 121 |
+
step 1190000 | 11613s | train_loss=0.0324
|
| 122 |
+
step 1200000 | 11778s | train_loss=0.0323 | fid=155.5721
|
| 123 |
+
step 1210000 | 11868s | train_loss=0.0321
|
| 124 |
+
step 1220000 | 11959s | train_loss=0.0323
|
| 125 |
+
step 1230000 | 12048s | train_loss=0.0321
|
| 126 |
+
step 1240000 | 12137s | train_loss=0.0309
|
| 127 |
+
step 1250000 | 12226s | train_loss=0.0311
|
| 128 |
+
step 1260000 | 12317s | train_loss=0.0310
|
| 129 |
+
step 1270000 | 12408s | train_loss=0.0309
|
| 130 |
+
step 1280000 | 12500s | train_loss=0.0305
|
| 131 |
+
step 1290000 | 12591s | train_loss=0.0308
|
| 132 |
+
step 1300000 | 12756s | train_loss=0.0306 | fid=155.6373
|
| 133 |
+
step 1310000 | 12846s | train_loss=0.0298
|
| 134 |
+
step 1320000 | 12937s | train_loss=0.0304
|
| 135 |
+
step 1330000 | 13023s | train_loss=0.0305
|
| 136 |
+
step 1340000 | 13112s | train_loss=0.0298
|
| 137 |
+
step 1350000 | 13203s | train_loss=0.0297
|
| 138 |
+
step 1360000 | 13294s | train_loss=0.0296
|
| 139 |
+
step 1370000 | 13386s | train_loss=0.0292
|
| 140 |
+
step 1380000 | 13477s | train_loss=0.0293
|
| 141 |
+
step 1390000 | 13568s | train_loss=0.0295
|
| 142 |
+
step 1400000 | 13732s | train_loss=0.0292 | fid=155.6435
|
| 143 |
+
step 1410000 | 13822s | train_loss=0.0288
|
| 144 |
+
step 1420000 | 13908s | train_loss=0.0288
|
| 145 |
+
step 1430000 | 13995s | train_loss=0.0285
|
| 146 |
+
step 1440000 | 14076s | train_loss=0.0281
|
| 147 |
+
step 1450000 | 14167s | train_loss=0.0279
|
| 148 |
+
step 1460000 | 14259s | train_loss=0.0283
|
| 149 |
+
step 1470000 | 14350s | train_loss=0.0280
|
| 150 |
+
step 1480000 | 14441s | train_loss=0.0271
|
| 151 |
+
step 1490000 | 14532s | train_loss=0.0276
|
| 152 |
+
step 1500000 | 14695s | train_loss=0.0274 | fid=155.4527
|
| 153 |
+
step 1510000 | 14783s | train_loss=0.0279
|
| 154 |
+
step 1520000 | 14871s | train_loss=0.0274
|
| 155 |
+
step 1530000 | 14963s | train_loss=0.0275
|
| 156 |
+
step 1540000 | 15051s | train_loss=0.0271
|
| 157 |
+
step 1550000 | 15142s | train_loss=0.0267
|
| 158 |
+
step 1560000 | 15233s | train_loss=0.0269
|
| 159 |
+
step 1570000 | 15325s | train_loss=0.0271
|
| 160 |
+
step 1580000 | 15416s | train_loss=0.0263
|
| 161 |
+
step 1590000 | 15507s | train_loss=0.0261
|
| 162 |
+
step 1600000 | 15672s | train_loss=0.0259 | fid=155.5546
|
| 163 |
+
step 1610000 | 15758s | train_loss=0.0266
|
| 164 |
+
step 1620000 | 15846s | train_loss=0.0261
|
| 165 |
+
step 1630000 | 15935s | train_loss=0.0261
|
| 166 |
+
step 1640000 | 16026s | train_loss=0.0254
|
| 167 |
+
step 1650000 | 16117s | train_loss=0.0255
|
| 168 |
+
step 1660000 | 16209s | train_loss=0.0258
|
| 169 |
+
step 1670000 | 16300s | train_loss=0.0257
|
| 170 |
+
step 1680000 | 16391s | train_loss=0.0261
|
| 171 |
+
step 1690000 | 16483s | train_loss=0.0254
|
| 172 |
+
step 1700000 | 16645s | train_loss=0.0250 | fid=155.5777
|
| 173 |
+
step 1710000 | 16729s | train_loss=0.0255
|
| 174 |
+
step 1720000 | 16818s | train_loss=0.0249
|
| 175 |
+
step 1730000 | 16897s | train_loss=0.0245
|
| 176 |
+
step 1740000 | 16988s | train_loss=0.0246
|
| 177 |
+
step 1750000 | 17079s | train_loss=0.0247
|
| 178 |
+
step 1760000 | 17171s | train_loss=0.0242
|
| 179 |
+
step 1770000 | 17262s | train_loss=0.0245
|
| 180 |
+
step 1780000 | 17353s | train_loss=0.0244
|
| 181 |
+
step 1790000 | 17445s | train_loss=0.0247
|
| 182 |
+
step 1800000 | 17607s | train_loss=0.0245 | fid=155.6740
|
| 183 |
+
step 1810000 | 17691s | train_loss=0.0241
|
| 184 |
+
step 1820000 | 17781s | train_loss=0.0246
|
| 185 |
+
step 1830000 | 17867s | train_loss=0.0246
|
| 186 |
+
step 1840000 | 17958s | train_loss=0.0240
|
| 187 |
+
step 1850000 | 18050s | train_loss=0.0241
|
| 188 |
+
step 1860000 | 18141s | train_loss=0.0239
|
| 189 |
+
step 1870000 | 18232s | train_loss=0.0235
|
| 190 |
+
step 1880000 | 18323s | train_loss=0.0236
|
| 191 |
+
step 1890000 | 18414s | train_loss=0.0233
|
| 192 |
+
step 1900000 | 18575s | train_loss=0.0237 | fid=155.4841
|
| 193 |
+
step 1910000 | 18667s | train_loss=0.0232
|
| 194 |
+
step 1920000 | 18754s | train_loss=0.0234
|
| 195 |
+
step 1930000 | 18845s | train_loss=0.0237
|
| 196 |
+
step 1940000 | 18936s | train_loss=0.0229
|
| 197 |
+
step 1950000 | 19028s | train_loss=0.0231
|
| 198 |
+
step 1960000 | 19119s | train_loss=0.0226
|
| 199 |
+
step 1970000 | 19210s | train_loss=0.0228
|
| 200 |
+
step 1980000 | 19302s | train_loss=0.0225
|
| 201 |
+
step 1990000 | 19391s | train_loss=0.0229
|
| 202 |
+
step 2000000 | 19545s | train_loss=0.0227 | fid=155.5070
|
| 203 |
+
step 2010000 | 19636s | train_loss=0.0226
|
| 204 |
+
step 2020000 | 19712s | train_loss=0.0223
|
| 205 |
+
step 2030000 | 19800s | train_loss=0.0228
|
| 206 |
+
step 2040000 | 19892s | train_loss=0.0224
|
| 207 |
+
step 2050000 | 19983s | train_loss=0.0220
|
| 208 |
+
step 2060000 | 20074s | train_loss=0.0222
|
| 209 |
+
step 2070000 | 20165s | train_loss=0.0220
|
| 210 |
+
step 2080000 | 20254s | train_loss=0.0221
|
| 211 |
+
step 2090000 | 20345s | train_loss=0.0220
|
| 212 |
+
step 2100000 | 20495s | train_loss=0.0217 | fid=155.3391
|
| 213 |
+
step 2110000 | 20587s | train_loss=0.0218
|
| 214 |
+
step 2120000 | 20669s | train_loss=0.0218
|
| 215 |
+
step 2130000 | 20755s | train_loss=0.0219
|
| 216 |
+
step 2140000 | 20846s | train_loss=0.0220
|
| 217 |
+
step 2150000 | 20938s | train_loss=0.0216
|
| 218 |
+
step 2160000 | 21028s | train_loss=0.0211
|
| 219 |
+
step 2170000 | 21120s | train_loss=0.0216
|
| 220 |
+
step 2180000 | 21209s | train_loss=0.0213
|
| 221 |
+
step 2190000 | 21301s | train_loss=0.0211
|
| 222 |
+
step 2200000 | 21450s | train_loss=0.0215 | fid=155.0475
|
| 223 |
+
step 2210000 | 21542s | train_loss=0.0211
|
| 224 |
+
step 2220000 | 21625s | train_loss=0.0211
|
| 225 |
+
step 2230000 | 21709s | train_loss=0.0210
|
| 226 |
+
step 2240000 | 21800s | train_loss=0.0206
|
| 227 |
+
step 2250000 | 21891s | train_loss=0.0211
|
| 228 |
+
step 2260000 | 21983s | train_loss=0.0208
|
| 229 |
+
step 2270000 | 22072s | train_loss=0.0205
|
| 230 |
+
step 2280000 | 22163s | train_loss=0.0206
|
| 231 |
+
step 2290000 | 22254s | train_loss=0.0210
|
| 232 |
+
step 2300000 | 22406s | train_loss=0.0201 | fid=155.2486
|
| 233 |
+
step 2310000 | 22497s | train_loss=0.0203
|
| 234 |
+
step 2320000 | 22583s | train_loss=0.0204
|
| 235 |
+
step 2330000 | 22664s | train_loss=0.0201
|
| 236 |
+
step 2340000 | 22755s | train_loss=0.0204
|
| 237 |
+
step 2350000 | 22847s | train_loss=0.0205
|
| 238 |
+
step 2360000 | 22936s | train_loss=0.0201
|
| 239 |
+
step 2370000 | 23027s | train_loss=0.0199
|
| 240 |
+
step 2380000 | 23118s | train_loss=0.0199
|
| 241 |
+
step 2390000 | 23209s | train_loss=0.0203
|
| 242 |
+
step 2400000 | 23360s | train_loss=0.0198 | fid=155.3335
|
| 243 |
+
step 2410000 | 23452s | train_loss=0.0199
|
| 244 |
+
step 2420000 | 23541s | train_loss=0.0198
|
| 245 |
+
step 2430000 | 23623s | train_loss=0.0195
|
| 246 |
+
step 2440000 | 23714s | train_loss=0.0195
|
| 247 |
+
step 2450000 | 23806s | train_loss=0.0196
|
| 248 |
+
step 2460000 | 23895s | train_loss=0.0201
|
| 249 |
+
step 2470000 | 23987s | train_loss=0.0194
|
| 250 |
+
step 2480000 | 24078s | train_loss=0.0196
|
| 251 |
+
step 2490000 | 24169s | train_loss=0.0190
|
| 252 |
+
step 2500000 | 24316s | train_loss=0.0197 | fid=155.7295
|
| 253 |
+
step 2510000 | 24408s | train_loss=0.0193
|
| 254 |
+
step 2520000 | 24496s | train_loss=0.0195
|
| 255 |
+
step 2530000 | 24577s | train_loss=0.0194
|
| 256 |
+
step 2540000 | 24668s | train_loss=0.0194
|
| 257 |
+
step 2550000 | 24757s | train_loss=0.0192
|
| 258 |
+
step 2560000 | 24849s | train_loss=0.0186
|
| 259 |
+
step 2570000 | 24940s | train_loss=0.0190
|
| 260 |
+
step 2580000 | 25031s | train_loss=0.0188
|
| 261 |
+
step 2590000 | 25122s | train_loss=0.0190
|
| 262 |
+
step 2600000 | 25273s | train_loss=0.0192 | fid=154.3313
|
| 263 |
+
step 2610000 | 25364s | train_loss=0.0188
|
| 264 |
+
step 2620000 | 25453s | train_loss=0.0183
|
| 265 |
+
step 2630000 | 25532s | train_loss=0.0186
|
| 266 |
+
step 2640000 | 25621s | train_loss=0.0186
|
| 267 |
+
step 2650000 | 25712s | train_loss=0.0185
|
| 268 |
+
step 2660000 | 25804s | train_loss=0.0184
|
| 269 |
+
step 2670000 | 25895s | train_loss=0.0186
|
| 270 |
+
step 2680000 | 25986s | train_loss=0.0181
|
| 271 |
+
step 2690000 | 26078s | train_loss=0.0185
|
| 272 |
+
step 2700000 | 26229s | train_loss=0.0184 | fid=155.2770
|
| 273 |
+
step 2710000 | 26320s | train_loss=0.0181
|
| 274 |
+
step 2720000 | 26410s | train_loss=0.0184
|
| 275 |
+
step 2730000 | 26497s | train_loss=0.0182
|
| 276 |
+
step 2740000 | 26586s | train_loss=0.0184
|
| 277 |
+
step 2750000 | 26678s | train_loss=0.0181
|
| 278 |
+
step 2760000 | 26769s | train_loss=0.0181
|
| 279 |
+
step 2770000 | 26861s | train_loss=0.0177
|
| 280 |
+
step 2780000 | 26952s | train_loss=0.0180
|
| 281 |
+
step 2790000 | 27043s | train_loss=0.0178
|
| 282 |
+
step 2800000 | 27205s | train_loss=0.0178 | fid=154.9913
|
| 283 |
+
step 2810000 | 27296s | train_loss=0.0177
|
| 284 |
+
step 2820000 | 27378s | train_loss=0.0179
|
| 285 |
+
step 2830000 | 27463s | train_loss=0.0179
|
| 286 |
+
step 2840000 | 27554s | train_loss=0.0174
|
| 287 |
+
step 2850000 | 27646s | train_loss=0.0178
|
| 288 |
+
step 2860000 | 27737s | train_loss=0.0175
|
| 289 |
+
step 2870000 | 27829s | train_loss=0.0172
|
| 290 |
+
step 2880000 | 27920s | train_loss=0.0173
|
| 291 |
+
step 2890000 | 28011s | train_loss=0.0177
|
| 292 |
+
step 2900000 | 28162s | train_loss=0.0174 | fid=155.2641
|
| 293 |
+
step 2910000 | 28254s | train_loss=0.0172
|
| 294 |
+
step 2920000 | 28335s | train_loss=0.0175
|
| 295 |
+
step 2930000 | 28418s | train_loss=0.0172
|
| 296 |
+
step 2940000 | 28509s | train_loss=0.0169
|
| 297 |
+
step 2950000 | 28600s | train_loss=0.0172
|
| 298 |
+
step 2960000 | 28692s | train_loss=0.0171
|
| 299 |
+
step 2970000 | 28783s | train_loss=0.0175
|
| 300 |
+
step 2980000 | 28875s | train_loss=0.0170
|
| 301 |
+
step 2990000 | 28966s | train_loss=0.0173
|
| 302 |
+
step 3000000 | 29115s | train_loss=0.0171 | fid=155.7164
|
| 303 |
+
step 3010000 | 29206s | train_loss=0.0168
|
| 304 |
+
step 3020000 | 29289s | train_loss=0.0166
|
| 305 |
+
step 3030000 | 29379s | train_loss=0.0164
|
| 306 |
+
step 3040000 | 29471s | train_loss=0.0166
|
| 307 |
+
step 3050000 | 29562s | train_loss=0.0166
|
| 308 |
+
step 3060000 | 29653s | train_loss=0.0167
|
| 309 |
+
step 3070000 | 29744s | train_loss=0.0171
|
| 310 |
+
step 3080000 | 29836s | train_loss=0.0162
|
| 311 |
+
step 3090000 | 29927s | train_loss=0.0167
|
| 312 |
+
step 3100000 | 30089s | train_loss=0.0164 | fid=155.2062
|
| 313 |
+
step 3110000 | 30178s | train_loss=0.0166
|
| 314 |
+
step 3120000 | 30266s | train_loss=0.0162
|
| 315 |
+
step 3130000 | 30356s | train_loss=0.0164
|
| 316 |
+
step 3140000 | 30448s | train_loss=0.0165
|
| 317 |
+
step 3150000 | 30539s | train_loss=0.0168
|
| 318 |
+
step 3160000 | 30630s | train_loss=0.0160
|
| 319 |
+
step 3170000 | 30722s | train_loss=0.0162
|
| 320 |
+
step 3180000 | 30814s | train_loss=0.0160
|
| 321 |
+
step 3190000 | 30902s | train_loss=0.0164
|
| 322 |
+
step 3200000 | 31059s | train_loss=0.0163 | fid=155.0998
|
| 323 |
+
step 3210000 | 31144s | train_loss=0.0160
|
| 324 |
+
step 3220000 | 31234s | train_loss=0.0159
|
| 325 |
+
step 3230000 | 31324s | train_loss=0.0162
|
| 326 |
+
step 3240000 | 31416s | train_loss=0.0161
|
| 327 |
+
step 3250000 | 31507s | train_loss=0.0157
|
| 328 |
+
step 3260000 | 31599s | train_loss=0.0159
|
| 329 |
+
step 3270000 | 31690s | train_loss=0.0157
|
| 330 |
+
step 3280000 | 31781s | train_loss=0.0160
|
| 331 |
+
step 3290000 | 31866s | train_loss=0.0154
|
| 332 |
+
step 3300000 | 32023s | train_loss=0.0161 | fid=154.9283
|
| 333 |
+
step 3310000 | 32106s | train_loss=0.0159
|
| 334 |
+
step 3320000 | 32198s | train_loss=0.0158
|
| 335 |
+
step 3330000 | 32287s | train_loss=0.0158
|
| 336 |
+
step 3340000 | 32379s | train_loss=0.0160
|
| 337 |
+
step 3350000 | 32470s | train_loss=0.0158
|
| 338 |
+
step 3360000 | 32561s | train_loss=0.0159
|
| 339 |
+
step 3370000 | 32653s | train_loss=0.0157
|
| 340 |
+
step 3380000 | 32744s | train_loss=0.0156
|
| 341 |
+
step 3390000 | 32826s | train_loss=0.0159
|
| 342 |
+
step 3400000 | 32988s | train_loss=0.0155 | fid=155.2211
|
| 343 |
+
step 3410000 | 33067s | train_loss=0.0159
|
| 344 |
+
step 3420000 | 33157s | train_loss=0.0154
|
| 345 |
+
step 3430000 | 33247s | train_loss=0.0155
|
| 346 |
+
step 3440000 | 33338s | train_loss=0.0153
|
| 347 |
+
step 3450000 | 33429s | train_loss=0.0153
|
| 348 |
+
step 3460000 | 33521s | train_loss=0.0152
|
| 349 |
+
step 3470000 | 33612s | train_loss=0.0151
|
| 350 |
+
step 3480000 | 33703s | train_loss=0.0153
|
| 351 |
+
step 3490000 | 33784s | train_loss=0.0150
|
| 352 |
+
step 3500000 | 33947s | train_loss=0.0152 | fid=155.0166
|
| 353 |
+
step 3510000 | 34029s | train_loss=0.0152
|
| 354 |
+
step 3520000 | 34121s | train_loss=0.0152
|
| 355 |
+
step 3530000 | 34211s | train_loss=0.0149
|
| 356 |
+
step 3540000 | 34303s | train_loss=0.0152
|
| 357 |
+
step 3550000 | 34394s | train_loss=0.0153
|
| 358 |
+
step 3560000 | 34485s | train_loss=0.0149
|
| 359 |
+
step 3570000 | 34577s | train_loss=0.0148
|
| 360 |
+
step 3580000 | 34668s | train_loss=0.0148
|
| 361 |
+
step 3590000 | 34749s | train_loss=0.0147
|
| 362 |
+
step 3600000 | 34913s | train_loss=0.0147 | fid=155.7835
|
| 363 |
+
step 3610000 | 34993s | train_loss=0.0150
|
| 364 |
+
step 3620000 | 35084s | train_loss=0.0148
|
| 365 |
+
step 3630000 | 35174s | train_loss=0.0147
|
| 366 |
+
step 3640000 | 35265s | train_loss=0.0150
|
| 367 |
+
step 3650000 | 35357s | train_loss=0.0148
|
| 368 |
+
step 3660000 | 35448s | train_loss=0.0147
|
| 369 |
+
step 3670000 | 35540s | train_loss=0.0142
|
| 370 |
+
step 3680000 | 35623s | train_loss=0.0146
|
| 371 |
+
step 3690000 | 35705s | train_loss=0.0145
|
| 372 |
+
step 3700000 | 35868s | train_loss=0.0145 | fid=154.9918
|
| 373 |
+
step 3710000 | 35947s | train_loss=0.0146
|
| 374 |
+
step 3720000 | 36039s | train_loss=0.0147
|
| 375 |
+
step 3730000 | 36129s | train_loss=0.0143
|
| 376 |
+
step 3740000 | 36220s | train_loss=0.0144
|
| 377 |
+
step 3750000 | 36312s | train_loss=0.0144
|
| 378 |
+
step 3760000 | 36403s | train_loss=0.0144
|
| 379 |
+
step 3770000 | 36495s | train_loss=0.0144
|
| 380 |
+
step 3780000 | 36579s | train_loss=0.0143
|
| 381 |
+
step 3790000 | 36661s | train_loss=0.0144
|
| 382 |
+
step 3800000 | 36825s | train_loss=0.0144 | fid=154.9274
|
| 383 |
+
step 3810000 | 36907s | train_loss=0.0144
|
| 384 |
+
step 3820000 | 36998s | train_loss=0.0142
|
| 385 |
+
step 3830000 | 37089s | train_loss=0.0143
|
| 386 |
+
step 3840000 | 37180s | train_loss=0.0144
|
| 387 |
+
step 3850000 | 37271s | train_loss=0.0141
|
| 388 |
+
step 3860000 | 37363s | train_loss=0.0139
|
| 389 |
+
step 3870000 | 37452s | train_loss=0.0142
|
| 390 |
+
step 3880000 | 37540s | train_loss=0.0144
|
| 391 |
+
step 3890000 | 37619s | train_loss=0.0139
|
| 392 |
+
step 3900000 | 37782s | train_loss=0.0137 | fid=155.0840
|
| 393 |
+
step 3910000 | 37863s | train_loss=0.0140
|
| 394 |
+
step 3920000 | 37955s | train_loss=0.0140
|
| 395 |
+
step 3930000 | 38045s | train_loss=0.0140
|
| 396 |
+
step 3940000 | 38136s | train_loss=0.0138
|
| 397 |
+
step 3950000 | 38228s | train_loss=0.0138
|
| 398 |
+
step 3960000 | 38318s | train_loss=0.0136
|
| 399 |
+
step 3970000 | 38409s | train_loss=0.0140
|
| 400 |
+
step 3980000 | 38499s | train_loss=0.0139
|
| 401 |
+
step 3990000 | 38578s | train_loss=0.0140
|
| 402 |
+
step 4000000 | 38743s | train_loss=0.0136 | fid=155.0248
|
| 403 |
+
step 4010000 | 38826s | train_loss=0.0136
|
| 404 |
+
step 4020000 | 38917s | train_loss=0.0139
|
| 405 |
+
step 4030000 | 39007s | train_loss=0.0142
|
| 406 |
+
step 4040000 | 39098s | train_loss=0.0138
|
| 407 |
+
step 4050000 | 39189s | train_loss=0.0139
|
| 408 |
+
step 4060000 | 39279s | train_loss=0.0139
|
| 409 |
+
step 4070000 | 39371s | train_loss=0.0133
|
| 410 |
+
step 4080000 | 39461s | train_loss=0.0136
|
| 411 |
+
step 4090000 | 39538s | train_loss=0.0137
|
| 412 |
+
step 4100000 | 39703s | train_loss=0.0135 | fid=154.7149
|
| 413 |
+
step 4110000 | 39783s | train_loss=0.0136
|
| 414 |
+
step 4120000 | 39874s | train_loss=0.0143
|
| 415 |
+
step 4130000 | 39964s | train_loss=0.0134
|
| 416 |
+
step 4140000 | 40055s | train_loss=0.0132
|
| 417 |
+
step 4150000 | 40145s | train_loss=0.0137
|
| 418 |
+
step 4160000 | 40236s | train_loss=0.0138
|
| 419 |
+
step 4170000 | 40327s | train_loss=0.0133
|
| 420 |
+
step 4180000 | 40419s | train_loss=0.0133
|
| 421 |
+
step 4190000 | 40495s | train_loss=0.0132
|
| 422 |
+
step 4200000 | 40660s | train_loss=0.0132 | fid=155.1374
|
| 423 |
+
step 4210000 | 40737s | train_loss=0.0131
|
| 424 |
+
step 4220000 | 40829s | train_loss=0.0133
|
| 425 |
+
step 4230000 | 40919s | train_loss=0.0132
|
| 426 |
+
step 4240000 | 41008s | train_loss=0.0132
|
| 427 |
+
step 4250000 | 41099s | train_loss=0.0134
|
| 428 |
+
step 4260000 | 41191s | train_loss=0.0129
|
| 429 |
+
step 4270000 | 41282s | train_loss=0.0134
|
| 430 |
+
step 4280000 | 41373s | train_loss=0.0132
|
| 431 |
+
step 4290000 | 41450s | train_loss=0.0134
|
| 432 |
+
step 4300000 | 41613s | train_loss=0.0133 | fid=155.7681
|
| 433 |
+
step 4310000 | 41690s | train_loss=0.0133
|
| 434 |
+
step 4320000 | 41782s | train_loss=0.0129
|
| 435 |
+
step 4330000 | 41872s | train_loss=0.0130
|
| 436 |
+
step 4340000 | 41962s | train_loss=0.0130
|
| 437 |
+
step 4350000 | 42053s | train_loss=0.0129
|
| 438 |
+
step 4360000 | 42144s | train_loss=0.0131
|
| 439 |
+
step 4370000 | 42236s | train_loss=0.0129
|
| 440 |
+
step 4380000 | 42327s | train_loss=0.0130
|
| 441 |
+
step 4390000 | 42403s | train_loss=0.0132
|
| 442 |
+
step 4400000 | 42563s | train_loss=0.0129 | fid=156.0403
|
| 443 |
+
step 4410000 | 42641s | train_loss=0.0128
|
| 444 |
+
step 4420000 | 42730s | train_loss=0.0127
|
| 445 |
+
step 4430000 | 42818s | train_loss=0.0130
|
| 446 |
+
step 4440000 | 42909s | train_loss=0.0126
|
| 447 |
+
step 4450000 | 43001s | train_loss=0.0129
|
| 448 |
+
step 4460000 | 43092s | train_loss=0.0130
|
| 449 |
+
step 4470000 | 43183s | train_loss=0.0129
|
| 450 |
+
step 4480000 | 43274s | train_loss=0.0136
|
| 451 |
+
step 4490000 | 43356s | train_loss=0.0128
|
| 452 |
+
step 4500000 | 43520s | train_loss=0.0127 | fid=155.6317
|
| 453 |
+
step 4510000 | 43603s | train_loss=0.0128
|
| 454 |
+
step 4520000 | 43690s | train_loss=0.0126
|
| 455 |
+
step 4530000 | 43780s | train_loss=0.0128
|
| 456 |
+
step 4540000 | 43870s | train_loss=0.0124
|
| 457 |
+
step 4550000 | 43961s | train_loss=0.0125
|
| 458 |
+
step 4560000 | 44053s | train_loss=0.0127
|
| 459 |
+
step 4570000 | 44144s | train_loss=0.0131
|
| 460 |
+
step 4580000 | 44236s | train_loss=0.0127
|
| 461 |
+
step 4590000 | 44317s | train_loss=0.0125
|
| 462 |
+
step 4600000 | 44481s | train_loss=0.0125 | fid=155.2310
|
| 463 |
+
step 4610000 | 44564s | train_loss=0.0124
|
| 464 |
+
step 4620000 | 44646s | train_loss=0.0127
|
| 465 |
+
step 4630000 | 44738s | train_loss=0.0127
|
| 466 |
+
step 4640000 | 44827s | train_loss=0.0126
|
| 467 |
+
step 4650000 | 44919s | train_loss=0.0127
|
| 468 |
+
step 4660000 | 45010s | train_loss=0.0123
|
| 469 |
+
step 4670000 | 45101s | train_loss=0.0126
|
| 470 |
+
step 4680000 | 45192s | train_loss=0.0123
|
| 471 |
+
step 4690000 | 45271s | train_loss=0.0124
|
| 472 |
+
step 4700000 | 45435s | train_loss=0.0128 | fid=155.7902
|
| 473 |
+
step 4710000 | 45516s | train_loss=0.0128
|
| 474 |
+
step 4720000 | 45583s | train_loss=0.0123
|
| 475 |
+
step 4730000 | 45657s | train_loss=0.0123
|
| 476 |
+
step 4740000 | 45728s | train_loss=0.0120
|
| 477 |
+
step 4750000 | 45802s | train_loss=0.0123
|
| 478 |
+
step 4760000 | 45876s | train_loss=0.0126
|
| 479 |
+
step 4770000 | 45949s | train_loss=0.0123
|
| 480 |
+
step 4780000 | 46021s | train_loss=0.0124
|
| 481 |
+
step 4790000 | 46078s | train_loss=0.0120
|
| 482 |
+
step 4800000 | 46211s | train_loss=0.0121 | fid=154.9014
|
| 483 |
+
step 4810000 | 46282s | train_loss=0.0122
|
| 484 |
+
step 4820000 | 46356s | train_loss=0.0119
|
| 485 |
+
step 4830000 | 46430s | train_loss=0.0124
|
| 486 |
+
step 4840000 | 46501s | train_loss=0.0125
|
| 487 |
+
step 4850000 | 46575s | train_loss=0.0124
|
| 488 |
+
step 4860000 | 46648s | train_loss=0.0122
|
| 489 |
+
step 4870000 | 46722s | train_loss=0.0120
|
| 490 |
+
step 4880000 | 46796s | train_loss=0.0123
|
| 491 |
+
step 4890000 | 46855s | train_loss=0.0121
|
| 492 |
+
step 4900000 | 46958s | train_loss=0.0120 | fid=155.4045
|
| 493 |
+
step 4910000 | 46998s | train_loss=0.0120
|
| 494 |
+
step 4920000 | 47036s | train_loss=0.0118
|
| 495 |
+
step 4930000 | 47074s | train_loss=0.0121
|
| 496 |
+
step 4940000 | 47109s | train_loss=0.0125
|
| 497 |
+
step 4950000 | 47147s | train_loss=0.0118
|
| 498 |
+
step 4960000 | 47185s | train_loss=0.0120
|
| 499 |
+
step 4970000 | 47223s | train_loss=0.0121
|
| 500 |
+
step 4980000 | 47261s | train_loss=0.0120
|
| 501 |
+
step 4990000 | 47299s | train_loss=0.0122
|
| 502 |
+
step 5000000 | 47372s | train_loss=0.0119 | fid=154.4299
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d200/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 45,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200/fid_retry.json
ADDED
|
@@ -0,0 +1,75 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"run_dir": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200",
|
| 3 |
+
"d_latent": 200,
|
| 4 |
+
"attempts": [
|
| 5 |
+
{
|
| 6 |
+
"attempt": 0,
|
| 7 |
+
"seed": 1045,
|
| 8 |
+
"step": 5000000,
|
| 9 |
+
"latent_finite": false,
|
| 10 |
+
"fid": null,
|
| 11 |
+
"error": "nonfinite_latents"
|
| 12 |
+
},
|
| 13 |
+
{
|
| 14 |
+
"attempt": 1,
|
| 15 |
+
"seed": 1046,
|
| 16 |
+
"step": 5000000,
|
| 17 |
+
"latent_finite": false,
|
| 18 |
+
"fid": null,
|
| 19 |
+
"error": "nonfinite_latents"
|
| 20 |
+
},
|
| 21 |
+
{
|
| 22 |
+
"attempt": 2,
|
| 23 |
+
"seed": 1047,
|
| 24 |
+
"step": 5000000,
|
| 25 |
+
"latent_finite": false,
|
| 26 |
+
"fid": null,
|
| 27 |
+
"error": "nonfinite_latents"
|
| 28 |
+
},
|
| 29 |
+
{
|
| 30 |
+
"attempt": 3,
|
| 31 |
+
"seed": 1048,
|
| 32 |
+
"step": 5000000,
|
| 33 |
+
"latent_finite": false,
|
| 34 |
+
"fid": null,
|
| 35 |
+
"error": "nonfinite_latents"
|
| 36 |
+
},
|
| 37 |
+
{
|
| 38 |
+
"attempt": 4,
|
| 39 |
+
"seed": 1049,
|
| 40 |
+
"step": 5000000,
|
| 41 |
+
"latent_finite": false,
|
| 42 |
+
"fid": null,
|
| 43 |
+
"error": "nonfinite_latents"
|
| 44 |
+
},
|
| 45 |
+
{
|
| 46 |
+
"attempt": 5,
|
| 47 |
+
"seed": 1050,
|
| 48 |
+
"step": 5000000,
|
| 49 |
+
"latent_finite": false,
|
| 50 |
+
"fid": null,
|
| 51 |
+
"error": "nonfinite_latents"
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"attempt": 6,
|
| 55 |
+
"seed": 1051,
|
| 56 |
+
"step": 5000000,
|
| 57 |
+
"latent_finite": true,
|
| 58 |
+
"pixel_finite": true,
|
| 59 |
+
"fid": 132.98561709967547,
|
| 60 |
+
"error": null
|
| 61 |
+
},
|
| 62 |
+
{
|
| 63 |
+
"attempt": 7,
|
| 64 |
+
"seed": 1052,
|
| 65 |
+
"step": 5000000,
|
| 66 |
+
"latent_finite": true,
|
| 67 |
+
"pixel_finite": true,
|
| 68 |
+
"fid": 133.24835054230078,
|
| 69 |
+
"error": null
|
| 70 |
+
}
|
| 71 |
+
],
|
| 72 |
+
"best_fid": 132.98561709967547,
|
| 73 |
+
"mean_fid": 133.11698382098814,
|
| 74 |
+
"n_success": 2
|
| 75 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200/fid_retry.log
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{"attempt": 6, "seed": 1051, "step": 5000000, "latent_finite": true, "pixel_finite": true, "fid": 132.98561709967547, "error": null}
|
| 2 |
+
{"attempt": 7, "seed": 1052, "step": 5000000, "latent_finite": true, "pixel_finite": true, "fid": 133.24835054230078, "error": null}
|
| 3 |
+
SUMMARY {"run_dir": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200", "d_latent": 200, "attempts": [{"attempt": 0, "seed": 1045, "step": 5000000, "latent_finite": false, "fid": null, "error": "nonfinite_latents"}, {"attempt": 1, "seed": 1046, "step": 5000000, "latent_finite": false, "fid": null, "error": "nonfinite_latents"}, {"attempt": 2, "seed": 1047, "step": 5000000, "latent_finite": false, "fid": null, "error": "nonfinite_latents"}, {"attempt": 3, "seed": 1048, "step": 5000000, "latent_finite": false, "fid": null, "error": "nonfinite_latents"}, {"attempt": 4, "seed": 1049, "step": 5000000, "latent_finite": false, "fid": null, "error": "nonfinite_latents"}, {"attempt": 5, "seed": 1050, "step": 5000000, "latent_finite": false, "fid": null, "error": "nonfinite_latents"}, {"attempt": 6, "seed": 1051, "step": 5000000, "latent_finite": true, "pixel_finite": true, "fid": 132.98561709967547, "error": null}, {"attempt": 7, "seed": 1052, "step": 5000000, "latent_finite": true, "pixel_finite": true, "fid": 133.24835054230078, "error": null}], "best_fid": 132.98561709967547, "mean_fid": 133.11698382098814, "n_success": 2}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d200/runner.log
ADDED
|
@@ -0,0 +1,502 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
FID real-side stats cached over 1000 diverse train images
|
| 2 |
+
step 1 | 73s | train_loss=1.0029 | fid=300.6978
|
| 3 |
+
step 10000 | 170s | train_loss=0.7587
|
| 4 |
+
step 20000 | 268s | train_loss=0.6520
|
| 5 |
+
step 30000 | 365s | train_loss=0.5869
|
| 6 |
+
step 40000 | 462s | train_loss=0.5293
|
| 7 |
+
step 50000 | 560s | train_loss=0.4659
|
| 8 |
+
step 60000 | 657s | train_loss=0.4046
|
| 9 |
+
step 70000 | 754s | train_loss=0.3498
|
| 10 |
+
step 80000 | 851s | train_loss=0.3052
|
| 11 |
+
step 90000 | 948s | train_loss=0.2651
|
| 12 |
+
step 100000 | 1118s | train_loss=0.2311 | fid=179.1970
|
| 13 |
+
step 110000 | 1215s | train_loss=0.2099
|
| 14 |
+
step 120000 | 1313s | train_loss=0.1947
|
| 15 |
+
step 130000 | 1410s | train_loss=0.1800
|
| 16 |
+
step 140000 | 1507s | train_loss=0.1750
|
| 17 |
+
step 150000 | 1605s | train_loss=0.1575
|
| 18 |
+
step 160000 | 1702s | train_loss=0.1508
|
| 19 |
+
step 170000 | 1800s | train_loss=0.1370
|
| 20 |
+
step 180000 | 1897s | train_loss=0.1376
|
| 21 |
+
step 190000 | 1989s | train_loss=0.1309
|
| 22 |
+
step 200000 | 2165s | train_loss=0.1343 | fid=227.9398
|
| 23 |
+
step 210000 | 2262s | train_loss=0.1336
|
| 24 |
+
step 220000 | 2360s | train_loss=0.1220
|
| 25 |
+
step 230000 | 2457s | train_loss=0.1146
|
| 26 |
+
step 240000 | 2554s | train_loss=0.1132
|
| 27 |
+
step 250000 | 2651s | train_loss=0.1019
|
| 28 |
+
step 260000 | 2749s | train_loss=0.0983
|
| 29 |
+
step 270000 | 2847s | train_loss=0.0975
|
| 30 |
+
step 280000 | 2939s | train_loss=0.1012
|
| 31 |
+
step 290000 | 3036s | train_loss=0.0994
|
| 32 |
+
step 300000 | 3209s | train_loss=0.0937 | fid=266.8161
|
| 33 |
+
step 310000 | 3306s | train_loss=0.0918
|
| 34 |
+
step 320000 | 3403s | train_loss=0.0940
|
| 35 |
+
step 330000 | 3501s | train_loss=0.0835
|
| 36 |
+
step 340000 | 3598s | train_loss=0.0873
|
| 37 |
+
step 350000 | 3695s | train_loss=0.0866
|
| 38 |
+
step 360000 | 3792s | train_loss=0.0800
|
| 39 |
+
step 370000 | 3886s | train_loss=0.0826
|
| 40 |
+
step 380000 | 3982s | train_loss=0.0752
|
| 41 |
+
step 390000 | 4079s | train_loss=0.0775
|
| 42 |
+
step 400000 | 4251s | train_loss=0.0744 | fid=282.5858
|
| 43 |
+
step 410000 | 4348s | train_loss=0.0791
|
| 44 |
+
step 420000 | 4445s | train_loss=0.0747
|
| 45 |
+
step 430000 | 4543s | train_loss=0.0696
|
| 46 |
+
step 440000 | 4640s | train_loss=0.0688
|
| 47 |
+
step 450000 | 4736s | train_loss=0.0751
|
| 48 |
+
step 460000 | 4832s | train_loss=0.0687
|
| 49 |
+
step 470000 | 4927s | train_loss=0.0683
|
| 50 |
+
step 480000 | 5024s | train_loss=0.0640
|
| 51 |
+
step 490000 | 5121s | train_loss=0.0648
|
| 52 |
+
step 500000 | 5289s | train_loss=0.0678 | fid=277.2592
|
| 53 |
+
step 510000 | 5384s | train_loss=0.0663
|
| 54 |
+
step 520000 | 5479s | train_loss=0.0643
|
| 55 |
+
step 530000 | 5574s | train_loss=0.0603
|
| 56 |
+
step 540000 | 5669s | train_loss=0.0645
|
| 57 |
+
step 550000 | 5766s | train_loss=0.0606
|
| 58 |
+
step 560000 | 5860s | train_loss=0.0633
|
| 59 |
+
step 570000 | 5957s | train_loss=0.0617
|
| 60 |
+
step 580000 | 6054s | train_loss=0.0635
|
| 61 |
+
step 590000 | 6152s | train_loss=0.0598
|
| 62 |
+
step 600000 | 6326s | train_loss=0.0661 | fid=256.3338
|
| 63 |
+
step 610000 | 6423s | train_loss=0.0674
|
| 64 |
+
step 620000 | 6521s | train_loss=0.0602
|
| 65 |
+
step 630000 | 6617s | train_loss=0.0600
|
| 66 |
+
step 640000 | 6715s | train_loss=0.0607
|
| 67 |
+
step 650000 | 6808s | train_loss=0.0593
|
| 68 |
+
step 660000 | 6905s | train_loss=0.0568
|
| 69 |
+
step 670000 | 7003s | train_loss=0.0557
|
| 70 |
+
step 680000 | 7101s | train_loss=0.0644
|
| 71 |
+
step 690000 | 7198s | train_loss=0.0588
|
| 72 |
+
step 700000 | 7373s | train_loss=0.0546 | fid=249.6294
|
| 73 |
+
step 710000 | 7471s | train_loss=0.0561
|
| 74 |
+
step 720000 | 7566s | train_loss=0.0569
|
| 75 |
+
step 730000 | 7663s | train_loss=0.0561
|
| 76 |
+
step 740000 | 7757s | train_loss=0.0519
|
| 77 |
+
step 750000 | 7854s | train_loss=0.0543
|
| 78 |
+
step 760000 | 7952s | train_loss=0.0540
|
| 79 |
+
step 770000 | 8049s | train_loss=0.0570
|
| 80 |
+
step 780000 | 8146s | train_loss=0.0532
|
| 81 |
+
step 790000 | 8244s | train_loss=0.0556
|
| 82 |
+
step 800000 | 8418s | train_loss=0.0543 | fid=233.8753
|
| 83 |
+
step 810000 | 8513s | train_loss=0.0485
|
| 84 |
+
step 820000 | 8611s | train_loss=0.0537
|
| 85 |
+
step 830000 | 8704s | train_loss=0.0502
|
| 86 |
+
step 840000 | 8802s | train_loss=0.0552
|
| 87 |
+
step 850000 | 8899s | train_loss=0.0525
|
| 88 |
+
step 860000 | 8996s | train_loss=0.0515
|
| 89 |
+
step 870000 | 9094s | train_loss=0.0521
|
| 90 |
+
step 880000 | 9191s | train_loss=0.0504
|
| 91 |
+
step 890000 | 9289s | train_loss=0.0467
|
| 92 |
+
step 900000 | 9459s | train_loss=0.0485 | fid=245.3049
|
| 93 |
+
step 910000 | 9556s | train_loss=0.0508
|
| 94 |
+
step 920000 | 9649s | train_loss=0.0599
|
| 95 |
+
step 930000 | 9746s | train_loss=0.0496
|
| 96 |
+
step 940000 | 9844s | train_loss=0.0514
|
| 97 |
+
step 950000 | 9942s | train_loss=0.0474
|
| 98 |
+
step 960000 | 10039s | train_loss=0.0519
|
| 99 |
+
step 970000 | 10137s | train_loss=0.0478
|
| 100 |
+
step 980000 | 10235s | train_loss=0.0485
|
| 101 |
+
step 990000 | 10330s | train_loss=0.0460
|
| 102 |
+
step 1000000 | 10504s | train_loss=0.0499 | fid=234.9598
|
| 103 |
+
step 1010000 | 10597s | train_loss=0.0463
|
| 104 |
+
step 1020000 | 10695s | train_loss=0.0468
|
| 105 |
+
step 1030000 | 10793s | train_loss=0.0478
|
| 106 |
+
step 1040000 | 10890s | train_loss=0.0458
|
| 107 |
+
step 1050000 | 10987s | train_loss=0.0482
|
| 108 |
+
step 1060000 | 11085s | train_loss=0.0490
|
| 109 |
+
step 1070000 | 11182s | train_loss=0.0639
|
| 110 |
+
step 1080000 | 11277s | train_loss=0.0461
|
| 111 |
+
step 1090000 | 11375s | train_loss=0.0492
|
| 112 |
+
step 1100000 | 11545s | train_loss=0.0480 | fid=232.3849
|
| 113 |
+
step 1110000 | 11642s | train_loss=0.0480
|
| 114 |
+
step 1120000 | 11739s | train_loss=0.0471
|
| 115 |
+
step 1130000 | 11837s | train_loss=0.0463
|
| 116 |
+
step 1140000 | 11934s | train_loss=0.0487
|
| 117 |
+
step 1150000 | 12032s | train_loss=0.0465
|
| 118 |
+
step 1160000 | 12129s | train_loss=0.0462
|
| 119 |
+
step 1170000 | 12225s | train_loss=0.0487
|
| 120 |
+
step 1180000 | 12322s | train_loss=0.0472
|
| 121 |
+
step 1190000 | 12420s | train_loss=0.0498
|
| 122 |
+
step 1200000 | 12591s | train_loss=0.0436 | fid=244.6573
|
| 123 |
+
step 1210000 | 12688s | train_loss=0.0448
|
| 124 |
+
step 1220000 | 12786s | train_loss=0.0468
|
| 125 |
+
step 1230000 | 12883s | train_loss=0.0435
|
| 126 |
+
step 1240000 | 12981s | train_loss=0.0486
|
| 127 |
+
step 1250000 | 13079s | train_loss=0.0432
|
| 128 |
+
step 1260000 | 13174s | train_loss=0.0494
|
| 129 |
+
step 1270000 | 13272s | train_loss=0.0439
|
| 130 |
+
step 1280000 | 13369s | train_loss=0.0464
|
| 131 |
+
step 1290000 | 13464s | train_loss=0.0455
|
| 132 |
+
step 1300000 | 13638s | train_loss=0.0433 | fid=235.8654
|
| 133 |
+
step 1310000 | 13736s | train_loss=0.0435
|
| 134 |
+
step 1320000 | 13833s | train_loss=0.0460
|
| 135 |
+
step 1330000 | 13931s | train_loss=0.0436
|
| 136 |
+
step 1340000 | 14028s | train_loss=0.0464
|
| 137 |
+
step 1350000 | 14124s | train_loss=0.0451
|
| 138 |
+
step 1360000 | 14222s | train_loss=0.0420
|
| 139 |
+
step 1370000 | 14319s | train_loss=0.0413
|
| 140 |
+
step 1380000 | 14409s | train_loss=0.0414
|
| 141 |
+
step 1390000 | 14501s | train_loss=0.0449
|
| 142 |
+
step 1400000 | 14671s | train_loss=0.0415 | fid=232.0209
|
| 143 |
+
step 1410000 | 14769s | train_loss=0.0416
|
| 144 |
+
step 1420000 | 14867s | train_loss=0.0411
|
| 145 |
+
step 1430000 | 14962s | train_loss=0.0485
|
| 146 |
+
step 1440000 | 15059s | train_loss=0.0459
|
| 147 |
+
step 1450000 | 15157s | train_loss=0.0423
|
| 148 |
+
step 1460000 | 15254s | train_loss=0.0402
|
| 149 |
+
step 1470000 | 15350s | train_loss=0.0447
|
| 150 |
+
step 1480000 | 15445s | train_loss=0.0408
|
| 151 |
+
step 1490000 | 15542s | train_loss=0.0463
|
| 152 |
+
step 1500000 | 15717s | train_loss=0.0428 | fid=220.5730
|
| 153 |
+
step 1510000 | 15815s | train_loss=0.0402
|
| 154 |
+
step 1520000 | 15911s | train_loss=0.0425
|
| 155 |
+
step 1530000 | 16008s | train_loss=0.0424
|
| 156 |
+
step 1540000 | 16106s | train_loss=0.0388
|
| 157 |
+
step 1550000 | 16204s | train_loss=0.0444
|
| 158 |
+
step 1560000 | 16300s | train_loss=0.0422
|
| 159 |
+
step 1570000 | 16386s | train_loss=0.0407
|
| 160 |
+
step 1580000 | 16483s | train_loss=0.0425
|
| 161 |
+
step 1590000 | 16581s | train_loss=0.0377
|
| 162 |
+
step 1600000 | 16786s | train_loss=0.0418 | fid=222.4675
|
| 163 |
+
step 1610000 | 16884s | train_loss=0.0398
|
| 164 |
+
step 1620000 | 16981s | train_loss=0.0434
|
| 165 |
+
step 1630000 | 17078s | train_loss=0.0392
|
| 166 |
+
step 1640000 | 17174s | train_loss=0.0393
|
| 167 |
+
step 1650000 | 17271s | train_loss=0.0398
|
| 168 |
+
step 1660000 | 17354s | train_loss=0.0458
|
| 169 |
+
step 1670000 | 17451s | train_loss=0.0400
|
| 170 |
+
step 1680000 | 17549s | train_loss=0.0377
|
| 171 |
+
step 1690000 | 17646s | train_loss=0.0411
|
| 172 |
+
step 1700000 | 17811s | train_loss=0.0377 | fid=210.7343
|
| 173 |
+
step 1710000 | 17908s | train_loss=0.0373
|
| 174 |
+
step 1720000 | 18006s | train_loss=0.0430
|
| 175 |
+
step 1730000 | 18101s | train_loss=0.0402
|
| 176 |
+
step 1740000 | 18199s | train_loss=0.0460
|
| 177 |
+
step 1750000 | 18295s | train_loss=0.0402
|
| 178 |
+
step 1760000 | 18391s | train_loss=0.0380
|
| 179 |
+
step 1770000 | 18488s | train_loss=0.0385
|
| 180 |
+
step 1780000 | 18586s | train_loss=0.0472
|
| 181 |
+
step 1790000 | 18681s | train_loss=0.0373
|
| 182 |
+
step 1800000 | 18856s | train_loss=0.0369 | fid=202.6979
|
| 183 |
+
step 1810000 | 18953s | train_loss=0.0402
|
| 184 |
+
step 1820000 | 19049s | train_loss=0.0379
|
| 185 |
+
step 1830000 | 19147s | train_loss=0.0391
|
| 186 |
+
step 1840000 | 19244s | train_loss=0.0375
|
| 187 |
+
step 1850000 | 19339s | train_loss=0.0354
|
| 188 |
+
step 1860000 | 19437s | train_loss=0.0372
|
| 189 |
+
step 1870000 | 19534s | train_loss=0.0414
|
| 190 |
+
step 1880000 | 19630s | train_loss=0.0370
|
| 191 |
+
step 1890000 | 19728s | train_loss=0.0388
|
| 192 |
+
step 1900000 | 19940s | train_loss=0.0387 | fid=203.2209
|
| 193 |
+
step 1910000 | 20036s | train_loss=0.0396
|
| 194 |
+
step 1920000 | 20133s | train_loss=0.0412
|
| 195 |
+
step 1930000 | 20221s | train_loss=0.0385
|
| 196 |
+
step 1940000 | 20317s | train_loss=0.0352
|
| 197 |
+
step 1950000 | 20415s | train_loss=0.0430
|
| 198 |
+
step 1960000 | 20510s | train_loss=0.0367
|
| 199 |
+
step 1970000 | 20608s | train_loss=0.0381
|
| 200 |
+
step 1980000 | 20705s | train_loss=0.0360
|
| 201 |
+
step 1990000 | 20803s | train_loss=0.0375
|
| 202 |
+
step 2000000 | 20996s | train_loss=0.0427 | fid=189.0424
|
| 203 |
+
step 2010000 | 21094s | train_loss=0.0363
|
| 204 |
+
step 2020000 | 21184s | train_loss=0.0386
|
| 205 |
+
step 2030000 | 21282s | train_loss=0.0341
|
| 206 |
+
step 2040000 | 21377s | train_loss=0.0357
|
| 207 |
+
step 2050000 | 21475s | train_loss=0.0382
|
| 208 |
+
step 2060000 | 21573s | train_loss=0.0354
|
| 209 |
+
step 2070000 | 21670s | train_loss=0.0409
|
| 210 |
+
step 2080000 | 21767s | train_loss=0.0355
|
| 211 |
+
step 2090000 | 21863s | train_loss=0.0356
|
| 212 |
+
step 2100000 | 22067s | train_loss=0.0385 | fid=192.4302
|
| 213 |
+
step 2110000 | 22153s | train_loss=0.0420
|
| 214 |
+
step 2120000 | 22246s | train_loss=0.0393
|
| 215 |
+
step 2130000 | 22343s | train_loss=0.0350
|
| 216 |
+
step 2140000 | 22440s | train_loss=0.0384
|
| 217 |
+
step 2150000 | 22538s | train_loss=0.0388
|
| 218 |
+
step 2160000 | 22636s | train_loss=0.0373
|
| 219 |
+
step 2170000 | 22733s | train_loss=0.0414
|
| 220 |
+
step 2180000 | 22829s | train_loss=0.0355
|
| 221 |
+
step 2190000 | 22927s | train_loss=0.0375
|
| 222 |
+
step 2200000 | 23119s | train_loss=0.0365 | fid=192.5769
|
| 223 |
+
step 2210000 | 23206s | train_loss=0.0364
|
| 224 |
+
step 2220000 | 23304s | train_loss=0.0380
|
| 225 |
+
step 2230000 | 23402s | train_loss=0.0337
|
| 226 |
+
step 2240000 | 23500s | train_loss=0.0321
|
| 227 |
+
step 2250000 | 23598s | train_loss=0.0360
|
| 228 |
+
step 2260000 | 23694s | train_loss=0.0350
|
| 229 |
+
step 2270000 | 23791s | train_loss=0.0359
|
| 230 |
+
step 2280000 | 23889s | train_loss=0.0336
|
| 231 |
+
step 2290000 | 23987s | train_loss=0.0421
|
| 232 |
+
step 2300000 | 24180s | train_loss=0.0336 | fid=192.0933
|
| 233 |
+
step 2310000 | 24278s | train_loss=0.0357
|
| 234 |
+
step 2320000 | 24377s | train_loss=0.0362
|
| 235 |
+
step 2330000 | 24474s | train_loss=0.0352
|
| 236 |
+
step 2340000 | 24572s | train_loss=0.0349
|
| 237 |
+
step 2350000 | 24668s | train_loss=0.0342
|
| 238 |
+
step 2360000 | 24766s | train_loss=0.0397
|
| 239 |
+
step 2370000 | 24863s | train_loss=0.0348
|
| 240 |
+
step 2380000 | 24961s | train_loss=0.0375
|
| 241 |
+
step 2390000 | 25056s | train_loss=0.0332
|
| 242 |
+
step 2400000 | 25232s | train_loss=0.0337 | fid=192.8534
|
| 243 |
+
step 2410000 | 25329s | train_loss=0.0338
|
| 244 |
+
step 2420000 | 25426s | train_loss=0.0325
|
| 245 |
+
step 2430000 | 25523s | train_loss=0.0354
|
| 246 |
+
step 2440000 | 25621s | train_loss=0.0341
|
| 247 |
+
step 2450000 | 25718s | train_loss=0.0337
|
| 248 |
+
step 2460000 | 25815s | train_loss=0.0489
|
| 249 |
+
step 2470000 | 25911s | train_loss=0.0336
|
| 250 |
+
step 2480000 | 26008s | train_loss=0.0353
|
| 251 |
+
step 2490000 | 26096s | train_loss=0.0345
|
| 252 |
+
step 2500000 | 26317s | train_loss=0.0355 | fid=186.6687
|
| 253 |
+
step 2510000 | 26415s | train_loss=0.0342
|
| 254 |
+
step 2520000 | 26511s | train_loss=0.0359
|
| 255 |
+
step 2530000 | 26608s | train_loss=0.0358
|
| 256 |
+
step 2540000 | 26706s | train_loss=0.0363
|
| 257 |
+
step 2550000 | 26801s | train_loss=0.0376
|
| 258 |
+
step 2560000 | 26899s | train_loss=0.0338
|
| 259 |
+
step 2570000 | 26996s | train_loss=0.0328
|
| 260 |
+
step 2580000 | 27091s | train_loss=0.0338
|
| 261 |
+
step 2590000 | 27189s | train_loss=0.0394
|
| 262 |
+
step 2600000 | 27381s | train_loss=0.0365 | fid=181.7482
|
| 263 |
+
step 2610000 | 27479s | train_loss=0.0356
|
| 264 |
+
step 2620000 | 27577s | train_loss=0.0371
|
| 265 |
+
step 2630000 | 27674s | train_loss=0.0354
|
| 266 |
+
step 2640000 | 27769s | train_loss=0.0357
|
| 267 |
+
step 2650000 | 27867s | train_loss=0.0337
|
| 268 |
+
step 2660000 | 27964s | train_loss=0.0352
|
| 269 |
+
step 2670000 | 28050s | train_loss=0.0331
|
| 270 |
+
step 2680000 | 28148s | train_loss=0.0331
|
| 271 |
+
step 2690000 | 28246s | train_loss=0.0317
|
| 272 |
+
step 2700000 | 28460s | train_loss=0.0317 | fid=180.6100
|
| 273 |
+
step 2710000 | 28557s | train_loss=0.0375
|
| 274 |
+
step 2720000 | 28653s | train_loss=0.0322
|
| 275 |
+
step 2730000 | 28751s | train_loss=0.0354
|
| 276 |
+
step 2740000 | 28848s | train_loss=0.0325
|
| 277 |
+
step 2750000 | 28943s | train_loss=0.0315
|
| 278 |
+
step 2760000 | 29030s | train_loss=0.0349
|
| 279 |
+
step 2770000 | 29128s | train_loss=0.0317
|
| 280 |
+
step 2780000 | 29225s | train_loss=0.0323
|
| 281 |
+
step 2790000 | 29321s | train_loss=0.0359
|
| 282 |
+
step 2800000 | 29531s | train_loss=0.0319 | fid=184.6122
|
| 283 |
+
step 2810000 | 29628s | train_loss=0.0344
|
| 284 |
+
step 2820000 | 29725s | train_loss=0.0370
|
| 285 |
+
step 2830000 | 29823s | train_loss=0.0355
|
| 286 |
+
step 2840000 | 29919s | train_loss=0.0311
|
| 287 |
+
step 2850000 | 30015s | train_loss=0.0324
|
| 288 |
+
step 2860000 | 30113s | train_loss=0.0326
|
| 289 |
+
step 2870000 | 30209s | train_loss=0.0350
|
| 290 |
+
step 2880000 | 30307s | train_loss=0.0428
|
| 291 |
+
step 2890000 | 30405s | train_loss=0.0305
|
| 292 |
+
step 2900000 | 30577s | train_loss=0.0309 | fid=171.3687
|
| 293 |
+
step 2910000 | 30675s | train_loss=0.0358
|
| 294 |
+
step 2920000 | 30772s | train_loss=0.0333
|
| 295 |
+
step 2930000 | 30869s | train_loss=0.0375
|
| 296 |
+
step 2940000 | 30965s | train_loss=0.0310
|
| 297 |
+
step 2950000 | 31062s | train_loss=0.0355
|
| 298 |
+
step 2960000 | 31158s | train_loss=0.0315
|
| 299 |
+
step 2970000 | 31256s | train_loss=0.0302
|
| 300 |
+
step 2980000 | 31353s | train_loss=0.0337
|
| 301 |
+
step 2990000 | 31449s | train_loss=0.0338
|
| 302 |
+
step 3000000 | 31659s | train_loss=0.0347 | fid=179.6911
|
| 303 |
+
step 3010000 | 31757s | train_loss=0.0324
|
| 304 |
+
step 3020000 | 31845s | train_loss=0.0318
|
| 305 |
+
step 3030000 | 31942s | train_loss=0.0292
|
| 306 |
+
step 3040000 | 32039s | train_loss=0.0326
|
| 307 |
+
step 3050000 | 32136s | train_loss=0.0332
|
| 308 |
+
step 3060000 | 32234s | train_loss=0.0337
|
| 309 |
+
step 3070000 | 32329s | train_loss=0.0319
|
| 310 |
+
step 3080000 | 32427s | train_loss=0.0335
|
| 311 |
+
step 3090000 | 32525s | train_loss=0.0301
|
| 312 |
+
step 3100000 | 32721s | train_loss=0.0316 | fid=188.1300
|
| 313 |
+
step 3110000 | 32810s | train_loss=0.0327
|
| 314 |
+
step 3120000 | 32907s | train_loss=0.0362
|
| 315 |
+
step 3130000 | 33003s | train_loss=0.0325
|
| 316 |
+
step 3140000 | 33100s | train_loss=0.0332
|
| 317 |
+
step 3150000 | 33196s | train_loss=0.0309
|
| 318 |
+
step 3160000 | 33293s | train_loss=0.0374
|
| 319 |
+
step 3170000 | 33391s | train_loss=0.0308
|
| 320 |
+
step 3180000 | 33488s | train_loss=0.0331
|
| 321 |
+
step 3190000 | 33586s | train_loss=0.0324
|
| 322 |
+
step 3200000 | 33768s | train_loss=0.0332 | fid=176.0478
|
| 323 |
+
step 3210000 | 33859s | train_loss=0.0303
|
| 324 |
+
step 3220000 | 33954s | train_loss=0.0321
|
| 325 |
+
step 3230000 | 34052s | train_loss=0.0302
|
| 326 |
+
step 3240000 | 34148s | train_loss=0.0362
|
| 327 |
+
step 3250000 | 34245s | train_loss=0.0324
|
| 328 |
+
step 3260000 | 34342s | train_loss=0.0305
|
| 329 |
+
step 3270000 | 34440s | train_loss=0.0342
|
| 330 |
+
step 3280000 | 34537s | train_loss=0.0324
|
| 331 |
+
step 3290000 | 34635s | train_loss=0.0319
|
| 332 |
+
step 3300000 | 34829s | train_loss=0.0338 | fid=172.0036
|
| 333 |
+
step 3310000 | 34926s | train_loss=0.0305
|
| 334 |
+
step 3320000 | 35022s | train_loss=0.0324
|
| 335 |
+
step 3330000 | 35119s | train_loss=0.0315
|
| 336 |
+
step 3340000 | 35217s | train_loss=0.0326
|
| 337 |
+
step 3350000 | 35314s | train_loss=0.0298
|
| 338 |
+
step 3360000 | 35412s | train_loss=0.0361
|
| 339 |
+
step 3370000 | 35509s | train_loss=0.0308
|
| 340 |
+
step 3380000 | 35607s | train_loss=0.0307
|
| 341 |
+
step 3390000 | 35706s | train_loss=0.0310
|
| 342 |
+
step 3400000 | 35875s | train_loss=0.0319 | fid=163.6520
|
| 343 |
+
step 3410000 | 35972s | train_loss=0.0322
|
| 344 |
+
step 3420000 | 36070s | train_loss=0.0313
|
| 345 |
+
step 3430000 | 36167s | train_loss=0.0326
|
| 346 |
+
step 3440000 | 36264s | train_loss=0.0277
|
| 347 |
+
step 3450000 | 36362s | train_loss=0.0316
|
| 348 |
+
step 3460000 | 36459s | train_loss=0.0315
|
| 349 |
+
step 3470000 | 36557s | train_loss=0.0309
|
| 350 |
+
step 3480000 | 36654s | train_loss=0.0313
|
| 351 |
+
step 3490000 | 36737s | train_loss=0.0334
|
| 352 |
+
step 3500000 | 36944s | train_loss=0.0318 | fid=160.5418
|
| 353 |
+
step 3510000 | 37042s | train_loss=0.0304
|
| 354 |
+
step 3520000 | 37139s | train_loss=0.0328
|
| 355 |
+
step 3530000 | 37237s | train_loss=0.0299
|
| 356 |
+
step 3540000 | 37334s | train_loss=0.0313
|
| 357 |
+
step 3550000 | 37432s | train_loss=0.0319
|
| 358 |
+
step 3560000 | 37529s | train_loss=0.0312
|
| 359 |
+
step 3570000 | 37626s | train_loss=0.0316
|
| 360 |
+
step 3580000 | 37711s | train_loss=0.0302
|
| 361 |
+
step 3590000 | 37809s | train_loss=0.0321
|
| 362 |
+
step 3600000 | 38022s | train_loss=0.0296 | fid=164.0121
|
| 363 |
+
step 3610000 | 38119s | train_loss=0.0313
|
| 364 |
+
step 3620000 | 38217s | train_loss=0.0309
|
| 365 |
+
step 3630000 | 38315s | train_loss=0.0295
|
| 366 |
+
step 3640000 | 38412s | train_loss=0.0303
|
| 367 |
+
step 3650000 | 38510s | train_loss=0.0305
|
| 368 |
+
step 3660000 | 38605s | train_loss=0.0304
|
| 369 |
+
step 3670000 | 38693s | train_loss=0.0311
|
| 370 |
+
step 3680000 | 38791s | train_loss=0.0328
|
| 371 |
+
step 3690000 | 38888s | train_loss=0.0303
|
| 372 |
+
step 3700000 | 39090s | train_loss=0.0310 | fid=164.9294
|
| 373 |
+
step 3710000 | 39188s | train_loss=0.0303
|
| 374 |
+
step 3720000 | 39285s | train_loss=0.0300
|
| 375 |
+
step 3730000 | 39383s | train_loss=0.0300
|
| 376 |
+
step 3740000 | 39479s | train_loss=0.0296
|
| 377 |
+
step 3750000 | 39574s | train_loss=0.0327
|
| 378 |
+
step 3760000 | 39663s | train_loss=0.0294
|
| 379 |
+
step 3770000 | 39761s | train_loss=0.0297
|
| 380 |
+
step 3780000 | 39858s | train_loss=0.0300
|
| 381 |
+
step 3790000 | 39955s | train_loss=0.0296
|
| 382 |
+
step 3800000 | 40145s | train_loss=0.0294 | fid=154.3482
|
| 383 |
+
step 3810000 | 40243s | train_loss=0.0286
|
| 384 |
+
step 3820000 | 40340s | train_loss=0.0298
|
| 385 |
+
step 3830000 | 40434s | train_loss=0.0320
|
| 386 |
+
step 3840000 | 40532s | train_loss=0.0276
|
| 387 |
+
step 3850000 | 40618s | train_loss=0.0318
|
| 388 |
+
step 3860000 | 40716s | train_loss=0.0307
|
| 389 |
+
step 3870000 | 40813s | train_loss=0.0279
|
| 390 |
+
step 3880000 | 40911s | train_loss=0.0310
|
| 391 |
+
step 3890000 | 41009s | train_loss=0.0288
|
| 392 |
+
step 3900000 | 41218s | train_loss=0.0309 | fid=154.8958
|
| 393 |
+
step 3910000 | 41313s | train_loss=0.0289
|
| 394 |
+
step 3920000 | 41410s | train_loss=0.0296
|
| 395 |
+
step 3930000 | 41507s | train_loss=0.0326
|
| 396 |
+
step 3940000 | 41593s | train_loss=0.0278
|
| 397 |
+
step 3950000 | 41691s | train_loss=0.0304
|
| 398 |
+
step 3960000 | 41788s | train_loss=0.0283
|
| 399 |
+
step 3970000 | 41886s | train_loss=0.0283
|
| 400 |
+
step 3980000 | 41984s | train_loss=0.0336
|
| 401 |
+
step 3990000 | 42081s | train_loss=0.0285
|
| 402 |
+
step 4000000 | 42296s | train_loss=0.0291 | fid=153.7553
|
| 403 |
+
step 4010000 | 42395s | train_loss=0.0296
|
| 404 |
+
step 4020000 | 42487s | train_loss=0.0301
|
| 405 |
+
step 4030000 | 42576s | train_loss=0.0290
|
| 406 |
+
step 4040000 | 42674s | train_loss=0.0273
|
| 407 |
+
step 4050000 | 42771s | train_loss=0.0278
|
| 408 |
+
step 4060000 | 42869s | train_loss=0.0303
|
| 409 |
+
step 4070000 | 42967s | train_loss=0.0284
|
| 410 |
+
step 4080000 | 43065s | train_loss=0.0295
|
| 411 |
+
step 4090000 | 43159s | train_loss=0.0297
|
| 412 |
+
step 4100000 | 43340s | train_loss=0.0266 | fid=141.8570
|
| 413 |
+
step 4110000 | 43437s | train_loss=0.0315
|
| 414 |
+
step 4120000 | 43526s | train_loss=0.0320
|
| 415 |
+
step 4130000 | 43624s | train_loss=0.0287
|
| 416 |
+
step 4140000 | 43722s | train_loss=0.0289
|
| 417 |
+
step 4150000 | 43820s | train_loss=0.0275
|
| 418 |
+
step 4160000 | 43917s | train_loss=0.0290
|
| 419 |
+
step 4170000 | 44014s | train_loss=0.0315
|
| 420 |
+
step 4180000 | 44109s | train_loss=0.0282
|
| 421 |
+
step 4190000 | 44206s | train_loss=0.0290
|
| 422 |
+
step 4200000 | 44418s | train_loss=0.0274 | fid=153.1675
|
| 423 |
+
step 4210000 | 44505s | train_loss=0.0275
|
| 424 |
+
step 4220000 | 44603s | train_loss=0.0279
|
| 425 |
+
step 4230000 | 44700s | train_loss=0.0269
|
| 426 |
+
step 4240000 | 44798s | train_loss=0.0258
|
| 427 |
+
step 4250000 | 44895s | train_loss=0.0286
|
| 428 |
+
step 4260000 | 44991s | train_loss=0.0283
|
| 429 |
+
step 4270000 | 45087s | train_loss=0.0312
|
| 430 |
+
step 4280000 | 45184s | train_loss=0.0282
|
| 431 |
+
step 4290000 | 45282s | train_loss=0.0331
|
| 432 |
+
step 4300000 | 45465s | train_loss=0.0287 | fid=140.3119
|
| 433 |
+
step 4310000 | 45563s | train_loss=0.0272
|
| 434 |
+
step 4320000 | 45660s | train_loss=0.0280
|
| 435 |
+
step 4330000 | 45757s | train_loss=0.0269
|
| 436 |
+
step 4340000 | 45851s | train_loss=0.0280
|
| 437 |
+
step 4350000 | 45928s | train_loss=0.0306
|
| 438 |
+
step 4360000 | 46006s | train_loss=0.0298
|
| 439 |
+
step 4370000 | 46085s | train_loss=0.0289
|
| 440 |
+
step 4380000 | 46164s | train_loss=0.0273
|
| 441 |
+
step 4390000 | 46242s | train_loss=0.0278
|
| 442 |
+
step 4400000 | 46319s | train_loss=0.0279 | fid=nan
|
| 443 |
+
step 4410000 | 46398s | train_loss=0.0277
|
| 444 |
+
step 4420000 | 46476s | train_loss=0.0273
|
| 445 |
+
step 4430000 | 46555s | train_loss=0.0306
|
| 446 |
+
step 4440000 | 46634s | train_loss=0.0311
|
| 447 |
+
step 4450000 | 46695s | train_loss=0.0269
|
| 448 |
+
step 4460000 | 46755s | train_loss=0.0319
|
| 449 |
+
step 4470000 | 46815s | train_loss=0.0286
|
| 450 |
+
step 4480000 | 46875s | train_loss=0.0279
|
| 451 |
+
step 4490000 | 46934s | train_loss=0.0272
|
| 452 |
+
step 4500000 | 47040s | train_loss=0.0286 | fid=137.6001
|
| 453 |
+
step 4510000 | 47100s | train_loss=0.0266
|
| 454 |
+
step 4520000 | 47160s | train_loss=0.0275
|
| 455 |
+
step 4530000 | 47219s | train_loss=0.0275
|
| 456 |
+
step 4540000 | 47279s | train_loss=0.0282
|
| 457 |
+
step 4550000 | 47339s | train_loss=0.0264
|
| 458 |
+
step 4560000 | 47399s | train_loss=0.0293
|
| 459 |
+
step 4570000 | 47458s | train_loss=0.0277
|
| 460 |
+
step 4580000 | 47518s | train_loss=0.0272
|
| 461 |
+
step 4590000 | 47569s | train_loss=0.0281
|
| 462 |
+
step 4600000 | 47642s | train_loss=0.0267 | fid=140.6708
|
| 463 |
+
step 4610000 | 47683s | train_loss=0.0284
|
| 464 |
+
step 4620000 | 47724s | train_loss=0.0256
|
| 465 |
+
step 4630000 | 47765s | train_loss=0.0268
|
| 466 |
+
step 4640000 | 47805s | train_loss=0.0264
|
| 467 |
+
step 4650000 | 47846s | train_loss=0.0273
|
| 468 |
+
step 4660000 | 47887s | train_loss=0.0269
|
| 469 |
+
step 4670000 | 47927s | train_loss=0.0304
|
| 470 |
+
step 4680000 | 47968s | train_loss=0.0274
|
| 471 |
+
step 4690000 | 48009s | train_loss=0.0278
|
| 472 |
+
step 4700000 | 48084s | train_loss=0.0275 | fid=138.3586
|
| 473 |
+
step 4710000 | 48125s | train_loss=0.0279
|
| 474 |
+
step 4720000 | 48166s | train_loss=0.0275
|
| 475 |
+
step 4730000 | 48206s | train_loss=0.0277
|
| 476 |
+
step 4740000 | 48247s | train_loss=0.0267
|
| 477 |
+
step 4750000 | 48288s | train_loss=0.0266
|
| 478 |
+
step 4760000 | 48329s | train_loss=0.0328
|
| 479 |
+
step 4770000 | 48370s | train_loss=0.0273
|
| 480 |
+
step 4780000 | 48411s | train_loss=0.0246
|
| 481 |
+
step 4790000 | 48451s | train_loss=0.0266
|
| 482 |
+
step 4800000 | 48524s | train_loss=0.0273 | fid=140.2433
|
| 483 |
+
step 4810000 | 48565s | train_loss=0.0299
|
| 484 |
+
step 4820000 | 48605s | train_loss=0.0302
|
| 485 |
+
step 4830000 | 48646s | train_loss=0.0280
|
| 486 |
+
step 4840000 | 48687s | train_loss=0.0277
|
| 487 |
+
step 4850000 | 48728s | train_loss=0.0260
|
| 488 |
+
step 4860000 | 48768s | train_loss=0.0288
|
| 489 |
+
step 4870000 | 48809s | train_loss=0.0279
|
| 490 |
+
step 4880000 | 48849s | train_loss=0.0296
|
| 491 |
+
step 4890000 | 48890s | train_loss=0.0283
|
| 492 |
+
step 4900000 | 48947s | train_loss=0.0284 | fid=nan
|
| 493 |
+
step 4910000 | 48982s | train_loss=0.0270
|
| 494 |
+
step 4920000 | 49023s | train_loss=0.0265
|
| 495 |
+
step 4930000 | 49064s | train_loss=0.0287
|
| 496 |
+
step 4940000 | 49104s | train_loss=0.0313
|
| 497 |
+
step 4950000 | 49145s | train_loss=0.0275
|
| 498 |
+
step 4960000 | 49186s | train_loss=0.0273
|
| 499 |
+
step 4970000 | 49226s | train_loss=0.0295
|
| 500 |
+
step 4980000 | 49267s | train_loss=0.0289
|
| 501 |
+
step 4990000 | 49308s | train_loss=0.0272
|
| 502 |
+
step 5000000 | 49365s | train_loss=0.0265 | fid=nan
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d220/config.json
ADDED
|
@@ -0,0 +1,25 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"vae_checkpoint": "vae_checkpoints/cifar10_resnet_beta005_d220/vae.pt",
|
| 3 |
+
"out": "results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d220",
|
| 4 |
+
"subset_path": "diagnostics/cifar10_diverse_1k_mean_plus_sd/subset_indices.json",
|
| 5 |
+
"n_train": 1000,
|
| 6 |
+
"total_steps": 5000000,
|
| 7 |
+
"eval_interval": 10000,
|
| 8 |
+
"mem_interval": 100000,
|
| 9 |
+
"fid_interval": 100000,
|
| 10 |
+
"fid_n_real": 1000,
|
| 11 |
+
"fid_n_gen": 10000,
|
| 12 |
+
"n_gen_samples": 10000,
|
| 13 |
+
"n_sde_steps": 500,
|
| 14 |
+
"batch_size": 512,
|
| 15 |
+
"hidden": 1024,
|
| 16 |
+
"depth": 5,
|
| 17 |
+
"optimizer": "sgd",
|
| 18 |
+
"lr": 0.001,
|
| 19 |
+
"momentum": 0.8,
|
| 20 |
+
"t_min": 0.01,
|
| 21 |
+
"t_max": 3.0,
|
| 22 |
+
"t_eval": 0.1,
|
| 23 |
+
"seed": 45,
|
| 24 |
+
"device": null
|
| 25 |
+
}
|
results/cifar10_beta005_diverse1k_bigmlp_sgd_lr001_m08_10k_5m/seed45/d220/metrics.jsonl
ADDED
|
The diff for this file is too large to render.
See raw diff
|
|
|