queue/cifar_grids.log

[2026-09-05 12:16:19] queue out/queue/cifar_grids.txt started, pid 14844
[2026-09-05 12:16:19] start [1]: .venv/bin/python scripts/build_cifar_grids.py --net resnet --device cpu --threads 4 --batch-size 32 --bank 256 --points 41 --probes 8 --bases 3 --time-bases 3 --kinds in_distribution --planes random --no-foreign --resume --cuts layer1.0 layer1.1 layer2.0 layer2.1 layer3.0 layer3.1 layer4.0 layer4.1
[12:16:21] bank (256, 3, 32, 32) from cifar100(cifar100)[train,n=256,seed=0,norm(0.5071,0.4867,0.4408|0.2675,0.2565,0.2761)]
[12:16:22] base types: in_distribution
[12:16:22] by depth at epoch 60: 8 cuts x 1 base types x 3 points x 1 planes, 41x41, jacobian probe (8 probes)
[12:33:39]   layer1.0: 3 written, 0 skipped (17.3 min)
[12:48:32]   layer1.1: 6 written, 0 skipped (32.2 min)
[13:01:08]   layer2.0: 9 written, 0 skipped (44.8 min)
[13:12:27]   layer2.1: 12 written, 0 skipped (56.1 min)
[13:21:21]   layer3.0: 15 written, 0 skipped (65.0 min)
[14:00:58]   layer3.1: 18 written, 0 skipped (104.6 min)
[14:03:51]   layer4.0: 21 written, 0 skipped (107.5 min)
[14:04:06]   layer4.1: 24 written, 0 skipped (107.7 min)
[14:04:06] by training time: 9 checkpoints x 1 base types at layer1.0
[14:22:40]   epoch 0: 27 written, 0 skipped (126.3 min)
[14:42:04]   epoch 1: 30 written, 0 skipped (145.7 min)
[15:01:07]   epoch 2: 33 written, 0 skipped (164.7 min)
[15:18:14]   epoch 5: 36 written, 0 skipped (181.9 min)
[15:35:48]   epoch 10: 39 written, 0 skipped (199.4 min)
[15:54:32]   epoch 20: 42 written, 0 skipped (218.2 min)
[16:12:34]   epoch 30: 45 written, 0 skipped (236.2 min)
[2026-09-05 16:24:41] exit [1]: status 143 after 14902s
[2026-09-05 16:24:41] stopping: command 1 failed