e2am: batch upload 13 files (322510 KB)

runs/tiny_imagenet/individual_methods/M3_grad_accum_only/best_model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:8fc1a0d7d929c6f64543f69128bb9df32ba8ffb3d0d7bbeafdb7ddbe81f1eb8d
+size 82657563

runs/tiny_imagenet/individual_methods/M3_grad_accum_only/checkpoints/checkpoint_last.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:fe05de4df1b4e8b78c76eda5461c40a59bea65a18128767af2174be4f864f1c6
+size 164532429

runs/tiny_imagenet/individual_methods/M3_grad_accum_only/config.yaml

@@ -0,0 +1,33 @@
+project_name: E2AM-ResNet50
+model_name: effnetv2_s
+dataset_name: tiny_imagenet
+output_root: /kaggle/working/E2AM-EffNetV2-TIN
+data_root: /kaggle/working/tiny-imagenet-200
+seed: 42
+num_epochs: 50
+batch_size: 64
+image_size: 32
+optimizer: sgd
+learning_rate: 0.1
+momentum: 0.9
+weight_decay: 0.0005
+warmup_epochs: 2
+amp_enabled: false
+cache_dataset: true
+gradient_accumulation_steps: 2
+l1_lambda: 0.0
+scheduler: none
+eag_enabled: false
+eag_threshold: 1.0e-07
+eag_patience: 5
+energy_sample_interval_sec: 1.0
+carbon_intensity_kg_per_kwh: 0.475
+milestone_push_every_epochs: 10
+heavy_push_every_epochs: 10
+force_rerun: false
+cleanup_local_after_complete: true
+cleanup_keep_summary: false
+model_input_size: 128
+method_group: individual_methods
+variant_name: M3_grad_accum_only
+method_id: M3

runs/tiny_imagenet/individual_methods/M3_grad_accum_only/final_model.pt

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5d1c336ce059f15bee1ff0eccb08925880fd89ab5b0fda8b4eb635728fab843d
+size 82658351

runs/tiny_imagenet/individual_methods/M3_grad_accum_only/history.csv

@@ -0,0 +1,5 @@
+epoch,train_loss,val_loss,train_accuracy,val_accuracy,val_accuracy_top5,f1_score,precision,recall,learning_rate,batch_size,effective_batch_size,amp_enabled,l1_lambda,gradient_accumulation_steps,epoch_time_sec,cumulative_time_sec,epoch_energy_j,cumulative_energy_j,epoch_co2_kg,cumulative_co2_kg,peak_vram_mb,eag_score,early_stop_counter,timestamp_utc
+0,5.005482255859375,4.5926317192077635,0.02703,0.0526,0.1909,0.03135266179248488,0.04007067640419772,0.05260000000000001,0.05,64,128,False,0.0,2,377.80437421798706,377.80437421798706,28441.73187383011,28441.73187383011,0.003752728511130362,0.003752728511130362,3383.490234375,0.0,0,2026-05-21T03:20:18Z
+1,4.362317757568359,4.007904506683349,0.08576,0.1319,0.3439,0.10432274662472028,0.1382022555121599,0.13190000000000002,0.1,64,128,False,0.0,2,387.4800651073456,765.2844393253326,28821.95054733767,57263.68242116778,0.00380289625277372,0.007555624763904082,3383.490234375,2.751375201680271e-06,0,2026-05-21T03:26:58Z
+2,3.888138208770752,3.9267655433654785,0.14469,0.142,0.366,0.11761322240973278,0.159809764657451,0.14200000000000002,0.1,64,128,False,0.0,2,377.27080273628235,1142.555242061615,28447.60582080869,85711.28824197646,0.003753503545801147,0.011309128309705228,3383.490234375,3.550386652437411e-07,0,2026-05-21T03:33:29Z
+3,3.6692502922058106,4.309495416259765,0.17884,0.1586,0.3911,0.13706032234828291,0.2079198671245737,0.1586,0.1,64,128,False,0.0,2,375.86719131469727,1518.4224333763123,28245.64352949287,113956.93177146933,0.0037268557434747534,0.015035984053179982,3383.490234375,5.877012496694227e-07,0,2026-05-21T03:39:59Z

runs/tiny_imagenet/individual_methods/M3_grad_accum_only/metrics_summary.json

@@ -0,0 +1,27 @@
+{
+  "run_key": "effnetv2_s::tiny_imagenet::individual_methods::M3_grad_accum_only",
+  "run_dir": "/kaggle/working/E2AM-EffNetV2-TIN/tiny_imagenet/individual_methods/M3_grad_accum_only",
+  "model_name": "effnetv2_s",
+  "dataset_name": "tiny_imagenet",
+  "method_group": "individual_methods",
+  "variant_name": "M3_grad_accum_only",
+  "method_id": "M3",
+  "num_epochs_planned": 50,
+  "num_epochs_run_so_far": 4,
+  "best_accuracy": 0.1586,
+  "latest_val_accuracy": 0.1586,
+  "latest_f1_score": 0.13706032234828291,
+  "total_time_sec_so_far": 1518.4224333763123,
+  "total_energy_j_so_far": 113956.93177146933,
+  "total_co2_kg_so_far": 0.015035984053179982,
+  "peak_vram_mb": 3383.490234375,
+  "amp_enabled": false,
+  "scheduler": "none",
+  "gradient_accumulation_steps": 2,
+  "l1_lambda": 0.0,
+  "eag_enabled": false,
+  "batch_size": 64,
+  "image_size": 32,
+  "status": "running",
+  "updated_utc": "2026-05-21T03:40:00Z"
+}