| { | |
| "best_metric": 0.7625570776255708, | |
| "best_model_checkpoint": "convnext-small-224-finetuned-piid/checkpoint-225", | |
| "epoch": 19.51219512195122, | |
| "eval_steps": 500, | |
| "global_step": 400, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.49, | |
| "learning_rate": 1.25e-05, | |
| "loss": 1.3915, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "learning_rate": 2.5e-05, | |
| "loss": 1.3405, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.98, | |
| "eval_accuracy": 0.4155251141552511, | |
| "eval_loss": 1.3200979232788086, | |
| "eval_runtime": 1.9477, | |
| "eval_samples_per_second": 112.44, | |
| "eval_steps_per_second": 14.376, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.46, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.2711, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.95, | |
| "learning_rate": 5e-05, | |
| "loss": 1.1715, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5707762557077626, | |
| "eval_loss": 1.1361732482910156, | |
| "eval_runtime": 2.0636, | |
| "eval_samples_per_second": 106.124, | |
| "eval_steps_per_second": 13.568, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.44, | |
| "learning_rate": 4.8611111111111115e-05, | |
| "loss": 1.0171, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.93, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.9231, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "eval_accuracy": 0.6438356164383562, | |
| "eval_loss": 0.925467848777771, | |
| "eval_runtime": 1.9933, | |
| "eval_samples_per_second": 109.87, | |
| "eval_steps_per_second": 14.047, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 3.41, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 0.7776, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.9, | |
| "learning_rate": 4.4444444444444447e-05, | |
| "loss": 0.7128, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6986301369863014, | |
| "eval_loss": 0.7557970881462097, | |
| "eval_runtime": 2.1609, | |
| "eval_samples_per_second": 101.346, | |
| "eval_steps_per_second": 12.958, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 4.305555555555556e-05, | |
| "loss": 0.6816, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.88, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.6204, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "eval_accuracy": 0.7534246575342466, | |
| "eval_loss": 0.705605685710907, | |
| "eval_runtime": 2.1297, | |
| "eval_samples_per_second": 102.833, | |
| "eval_steps_per_second": 13.148, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 5.37, | |
| "learning_rate": 4.027777777777778e-05, | |
| "loss": 0.5559, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.5322, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7397260273972602, | |
| "eval_loss": 0.6609848141670227, | |
| "eval_runtime": 2.1711, | |
| "eval_samples_per_second": 100.869, | |
| "eval_steps_per_second": 12.896, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 6.34, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 0.4642, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.83, | |
| "learning_rate": 3.611111111111111e-05, | |
| "loss": 0.4403, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.98, | |
| "eval_accuracy": 0.7442922374429224, | |
| "eval_loss": 0.6638585329055786, | |
| "eval_runtime": 2.0236, | |
| "eval_samples_per_second": 108.225, | |
| "eval_steps_per_second": 13.837, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 7.32, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.4919, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 7.8, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.4388, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.730593607305936, | |
| "eval_loss": 0.647175669670105, | |
| "eval_runtime": 2.0571, | |
| "eval_samples_per_second": 106.462, | |
| "eval_steps_per_second": 13.612, | |
| "step": 164 | |
| }, | |
| { | |
| "epoch": 8.29, | |
| "learning_rate": 3.194444444444444e-05, | |
| "loss": 0.3633, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 8.78, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.3901, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.98, | |
| "eval_accuracy": 0.7351598173515982, | |
| "eval_loss": 0.6684472560882568, | |
| "eval_runtime": 2.0915, | |
| "eval_samples_per_second": 104.711, | |
| "eval_steps_per_second": 13.388, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 9.27, | |
| "learning_rate": 2.916666666666667e-05, | |
| "loss": 0.3475, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 9.76, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.4202, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7397260273972602, | |
| "eval_loss": 0.5934084057807922, | |
| "eval_runtime": 2.2374, | |
| "eval_samples_per_second": 97.88, | |
| "eval_steps_per_second": 12.514, | |
| "step": 205 | |
| }, | |
| { | |
| "epoch": 10.24, | |
| "learning_rate": 2.6388888888888892e-05, | |
| "loss": 0.3367, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 10.73, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.3784, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 10.98, | |
| "eval_accuracy": 0.7625570776255708, | |
| "eval_loss": 0.5651447772979736, | |
| "eval_runtime": 2.0962, | |
| "eval_samples_per_second": 104.476, | |
| "eval_steps_per_second": 13.358, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 11.22, | |
| "learning_rate": 2.361111111111111e-05, | |
| "loss": 0.2953, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 11.71, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.2973, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7579908675799086, | |
| "eval_loss": 0.6438630819320679, | |
| "eval_runtime": 2.0644, | |
| "eval_samples_per_second": 106.084, | |
| "eval_steps_per_second": 13.563, | |
| "step": 246 | |
| }, | |
| { | |
| "epoch": 12.2, | |
| "learning_rate": 2.0833333333333336e-05, | |
| "loss": 0.2978, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 12.68, | |
| "learning_rate": 1.9444444444444445e-05, | |
| "loss": 0.3614, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 12.98, | |
| "eval_accuracy": 0.7534246575342466, | |
| "eval_loss": 0.5844311714172363, | |
| "eval_runtime": 2.0548, | |
| "eval_samples_per_second": 106.581, | |
| "eval_steps_per_second": 13.627, | |
| "step": 266 | |
| }, | |
| { | |
| "epoch": 13.17, | |
| "learning_rate": 1.8055555555555555e-05, | |
| "loss": 0.3089, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 13.66, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.2795, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.730593607305936, | |
| "eval_loss": 0.6014636158943176, | |
| "eval_runtime": 2.1523, | |
| "eval_samples_per_second": 101.752, | |
| "eval_steps_per_second": 13.009, | |
| "step": 287 | |
| }, | |
| { | |
| "epoch": 14.15, | |
| "learning_rate": 1.527777777777778e-05, | |
| "loss": 0.3327, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 14.63, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.2825, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 14.98, | |
| "eval_accuracy": 0.7625570776255708, | |
| "eval_loss": 0.6030529141426086, | |
| "eval_runtime": 2.0651, | |
| "eval_samples_per_second": 106.049, | |
| "eval_steps_per_second": 13.559, | |
| "step": 307 | |
| }, | |
| { | |
| "epoch": 15.12, | |
| "learning_rate": 1.25e-05, | |
| "loss": 0.2496, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 15.61, | |
| "learning_rate": 1.1111111111111112e-05, | |
| "loss": 0.2364, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7534246575342466, | |
| "eval_loss": 0.6248753666877747, | |
| "eval_runtime": 2.0546, | |
| "eval_samples_per_second": 106.592, | |
| "eval_steps_per_second": 13.628, | |
| "step": 328 | |
| }, | |
| { | |
| "epoch": 16.1, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.2543, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 16.59, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.2162, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 16.98, | |
| "eval_accuracy": 0.7625570776255708, | |
| "eval_loss": 0.6248474717140198, | |
| "eval_runtime": 2.1992, | |
| "eval_samples_per_second": 99.582, | |
| "eval_steps_per_second": 12.732, | |
| "step": 348 | |
| }, | |
| { | |
| "epoch": 17.07, | |
| "learning_rate": 6.944444444444445e-06, | |
| "loss": 0.2588, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 17.56, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.2455, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7488584474885844, | |
| "eval_loss": 0.6152719855308533, | |
| "eval_runtime": 2.0686, | |
| "eval_samples_per_second": 105.868, | |
| "eval_steps_per_second": 13.536, | |
| "step": 369 | |
| }, | |
| { | |
| "epoch": 18.05, | |
| "learning_rate": 4.166666666666667e-06, | |
| "loss": 0.3108, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 18.54, | |
| "learning_rate": 2.777777777777778e-06, | |
| "loss": 0.2314, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 18.98, | |
| "eval_accuracy": 0.7579908675799086, | |
| "eval_loss": 0.6113206148147583, | |
| "eval_runtime": 2.1977, | |
| "eval_samples_per_second": 99.648, | |
| "eval_steps_per_second": 12.74, | |
| "step": 389 | |
| }, | |
| { | |
| "epoch": 19.02, | |
| "learning_rate": 1.388888888888889e-06, | |
| "loss": 0.2958, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 19.51, | |
| "learning_rate": 0.0, | |
| "loss": 0.248, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.51, | |
| "eval_accuracy": 0.7579908675799086, | |
| "eval_loss": 0.6130588054656982, | |
| "eval_runtime": 2.18, | |
| "eval_samples_per_second": 100.46, | |
| "eval_steps_per_second": 12.844, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 19.51, | |
| "step": 400, | |
| "total_flos": 5.6921285921491354e+17, | |
| "train_loss": 0.5017214328050613, | |
| "train_runtime": 493.2709, | |
| "train_samples_per_second": 26.476, | |
| "train_steps_per_second": 0.811 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 400, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 5.6921285921491354e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |