| { | |
| "best_metric": 0.8943801522254944, | |
| "best_model_checkpoint": "./vit-base-renovation/checkpoint-200", | |
| "epoch": 8.0, | |
| "global_step": 456, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.18, | |
| "learning_rate": 0.0001956140350877193, | |
| "loss": 0.9993, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.35, | |
| "learning_rate": 0.0001912280701754386, | |
| "loss": 0.9628, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.53, | |
| "learning_rate": 0.00018684210526315792, | |
| "loss": 1.058, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.7, | |
| "learning_rate": 0.0001824561403508772, | |
| "loss": 0.9951, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.88, | |
| "learning_rate": 0.0001780701754385965, | |
| "loss": 1.0107, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 1.05, | |
| "learning_rate": 0.0001736842105263158, | |
| "loss": 1.0481, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 0.0001692982456140351, | |
| "loss": 0.9934, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.4, | |
| "learning_rate": 0.0001649122807017544, | |
| "loss": 0.9848, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.58, | |
| "learning_rate": 0.0001605263157894737, | |
| "loss": 0.8844, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "learning_rate": 0.00015614035087719297, | |
| "loss": 0.8483, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.75, | |
| "eval_accuracy": 0.5445544554455446, | |
| "eval_loss": 0.9965408444404602, | |
| "eval_runtime": 4.6358, | |
| "eval_samples_per_second": 21.787, | |
| "eval_steps_per_second": 2.804, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 0.0001517543859649123, | |
| "loss": 0.9321, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 2.11, | |
| "learning_rate": 0.00014736842105263158, | |
| "loss": 0.7607, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.28, | |
| "learning_rate": 0.00014298245614035088, | |
| "loss": 0.6632, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.46, | |
| "learning_rate": 0.00013859649122807018, | |
| "loss": 0.7359, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.63, | |
| "learning_rate": 0.00013421052631578948, | |
| "loss": 0.7393, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.81, | |
| "learning_rate": 0.0001298245614035088, | |
| "loss": 0.6817, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.98, | |
| "learning_rate": 0.00012543859649122806, | |
| "loss": 0.6913, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 3.16, | |
| "learning_rate": 0.00012149122807017544, | |
| "loss": 0.5017, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 3.33, | |
| "learning_rate": 0.00011710526315789474, | |
| "loss": 0.3448, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "learning_rate": 0.00011271929824561404, | |
| "loss": 0.3474, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.51, | |
| "eval_accuracy": 0.6831683168316832, | |
| "eval_loss": 0.8943801522254944, | |
| "eval_runtime": 4.5642, | |
| "eval_samples_per_second": 22.129, | |
| "eval_steps_per_second": 2.848, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.68, | |
| "learning_rate": 0.00010833333333333333, | |
| "loss": 0.3598, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 0.00010394736842105264, | |
| "loss": 0.3214, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 4.04, | |
| "learning_rate": 9.956140350877193e-05, | |
| "loss": 0.2831, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 4.21, | |
| "learning_rate": 9.517543859649123e-05, | |
| "loss": 0.1586, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.39, | |
| "learning_rate": 9.078947368421054e-05, | |
| "loss": 0.0929, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.56, | |
| "learning_rate": 8.640350877192982e-05, | |
| "loss": 0.1087, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 8.201754385964913e-05, | |
| "loss": 0.1019, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "learning_rate": 7.763157894736843e-05, | |
| "loss": 0.0528, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 5.09, | |
| "learning_rate": 7.324561403508772e-05, | |
| "loss": 0.1051, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "learning_rate": 6.885964912280702e-05, | |
| "loss": 0.0328, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.26, | |
| "eval_accuracy": 0.6633663366336634, | |
| "eval_loss": 1.158254861831665, | |
| "eval_runtime": 4.4598, | |
| "eval_samples_per_second": 22.647, | |
| "eval_steps_per_second": 2.915, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 5.44, | |
| "learning_rate": 6.447368421052632e-05, | |
| "loss": 0.0686, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 5.61, | |
| "learning_rate": 6.0087719298245616e-05, | |
| "loss": 0.0358, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.79, | |
| "learning_rate": 5.570175438596491e-05, | |
| "loss": 0.0255, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.96, | |
| "learning_rate": 5.131578947368422e-05, | |
| "loss": 0.0401, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 6.14, | |
| "learning_rate": 4.6929824561403515e-05, | |
| "loss": 0.0217, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 6.32, | |
| "learning_rate": 4.254385964912281e-05, | |
| "loss": 0.0208, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 6.49, | |
| "learning_rate": 3.815789473684211e-05, | |
| "loss": 0.0203, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 6.67, | |
| "learning_rate": 3.377192982456141e-05, | |
| "loss": 0.0196, | |
| "step": 380 | |
| }, | |
| { | |
| "epoch": 6.84, | |
| "learning_rate": 2.9385964912280706e-05, | |
| "loss": 0.0176, | |
| "step": 390 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.0176, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.02, | |
| "eval_accuracy": 0.6831683168316832, | |
| "eval_loss": 1.0844857692718506, | |
| "eval_runtime": 4.6632, | |
| "eval_samples_per_second": 21.659, | |
| "eval_steps_per_second": 2.788, | |
| "step": 400 | |
| }, | |
| { | |
| "epoch": 7.19, | |
| "learning_rate": 2.06140350877193e-05, | |
| "loss": 0.0168, | |
| "step": 410 | |
| }, | |
| { | |
| "epoch": 7.37, | |
| "learning_rate": 1.62280701754386e-05, | |
| "loss": 0.0165, | |
| "step": 420 | |
| }, | |
| { | |
| "epoch": 7.54, | |
| "learning_rate": 1.1842105263157895e-05, | |
| "loss": 0.0161, | |
| "step": 430 | |
| }, | |
| { | |
| "epoch": 7.72, | |
| "learning_rate": 7.456140350877193e-06, | |
| "loss": 0.0163, | |
| "step": 440 | |
| }, | |
| { | |
| "epoch": 7.89, | |
| "learning_rate": 3.070175438596491e-06, | |
| "loss": 0.0162, | |
| "step": 450 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "step": 456, | |
| "total_flos": 5.6352679914450125e+17, | |
| "train_loss": 0.3986694132745789, | |
| "train_runtime": 506.3898, | |
| "train_samples_per_second": 14.36, | |
| "train_steps_per_second": 0.9 | |
| } | |
| ], | |
| "max_steps": 456, | |
| "num_train_epochs": 8, | |
| "total_flos": 5.6352679914450125e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |