| { | |
| "best_metric": 0.8620689655172413, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-noh\\checkpoint-92", | |
| "epoch": 9.577777777777778, | |
| "eval_steps": 500, | |
| "global_step": 220, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.4444444444444444, | |
| "grad_norm": 12.336226463317871, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.7914, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.8888888888888888, | |
| "grad_norm": 8.176236152648926, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.5322, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.7619047619047619, | |
| "eval_loss": 0.46465402841567993, | |
| "eval_runtime": 64.6197, | |
| "eval_samples_per_second": 9.424, | |
| "eval_steps_per_second": 0.604, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.3111111111111111, | |
| "grad_norm": 12.591184616088867, | |
| "learning_rate": 4.797979797979798e-05, | |
| "loss": 0.4748, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.7555555555555555, | |
| "grad_norm": 11.170787811279297, | |
| "learning_rate": 4.545454545454546e-05, | |
| "loss": 0.4535, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.819376026272578, | |
| "eval_loss": 0.4359256625175476, | |
| "eval_runtime": 64.0615, | |
| "eval_samples_per_second": 9.506, | |
| "eval_steps_per_second": 0.609, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.1777777777777776, | |
| "grad_norm": 13.398783683776855, | |
| "learning_rate": 4.292929292929293e-05, | |
| "loss": 0.323, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.6222222222222222, | |
| "grad_norm": 14.519952774047852, | |
| "learning_rate": 4.0404040404040405e-05, | |
| "loss": 0.3854, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8538587848932676, | |
| "eval_loss": 0.3514404892921448, | |
| "eval_runtime": 64.0412, | |
| "eval_samples_per_second": 9.51, | |
| "eval_steps_per_second": 0.609, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.0444444444444443, | |
| "grad_norm": 9.444785118103027, | |
| "learning_rate": 3.787878787878788e-05, | |
| "loss": 0.3279, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.488888888888889, | |
| "grad_norm": 25.448450088500977, | |
| "learning_rate": 3.535353535353535e-05, | |
| "loss": 0.3003, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 3.9333333333333336, | |
| "grad_norm": 14.186548233032227, | |
| "learning_rate": 3.282828282828283e-05, | |
| "loss": 0.302, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8620689655172413, | |
| "eval_loss": 0.4349357783794403, | |
| "eval_runtime": 64.2242, | |
| "eval_samples_per_second": 9.482, | |
| "eval_steps_per_second": 0.607, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.355555555555555, | |
| "grad_norm": 10.739775657653809, | |
| "learning_rate": 3.0303030303030306e-05, | |
| "loss": 0.2932, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "grad_norm": 25.793798446655273, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2571, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8095238095238095, | |
| "eval_loss": 0.5112400054931641, | |
| "eval_runtime": 64.0219, | |
| "eval_samples_per_second": 9.512, | |
| "eval_steps_per_second": 0.609, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 5.222222222222222, | |
| "grad_norm": 14.213773727416992, | |
| "learning_rate": 2.5252525252525256e-05, | |
| "loss": 0.2101, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 5.666666666666667, | |
| "grad_norm": 48.91055679321289, | |
| "learning_rate": 2.272727272727273e-05, | |
| "loss": 0.2104, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.825944170771757, | |
| "eval_loss": 0.445329487323761, | |
| "eval_runtime": 64.6759, | |
| "eval_samples_per_second": 9.416, | |
| "eval_steps_per_second": 0.603, | |
| "step": 138 | |
| }, | |
| { | |
| "epoch": 6.088888888888889, | |
| "grad_norm": 17.380577087402344, | |
| "learning_rate": 2.0202020202020203e-05, | |
| "loss": 0.2352, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 6.533333333333333, | |
| "grad_norm": 14.665043830871582, | |
| "learning_rate": 1.7676767676767676e-05, | |
| "loss": 0.2105, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 6.977777777777778, | |
| "grad_norm": 16.076261520385742, | |
| "learning_rate": 1.5151515151515153e-05, | |
| "loss": 0.1702, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.7832512315270936, | |
| "eval_loss": 0.554968535900116, | |
| "eval_runtime": 64.6137, | |
| "eval_samples_per_second": 9.425, | |
| "eval_steps_per_second": 0.604, | |
| "step": 161 | |
| }, | |
| { | |
| "epoch": 7.4, | |
| "grad_norm": 20.720272064208984, | |
| "learning_rate": 1.2626262626262628e-05, | |
| "loss": 0.1453, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 7.844444444444444, | |
| "grad_norm": 15.213704109191895, | |
| "learning_rate": 1.0101010101010101e-05, | |
| "loss": 0.1682, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7947454844006568, | |
| "eval_loss": 0.5312520265579224, | |
| "eval_runtime": 65.466, | |
| "eval_samples_per_second": 9.303, | |
| "eval_steps_per_second": 0.596, | |
| "step": 184 | |
| }, | |
| { | |
| "epoch": 8.266666666666667, | |
| "grad_norm": 32.07905197143555, | |
| "learning_rate": 7.5757575757575764e-06, | |
| "loss": 0.123, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 8.71111111111111, | |
| "grad_norm": 12.448338508605957, | |
| "learning_rate": 5.050505050505051e-06, | |
| "loss": 0.136, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.8275862068965517, | |
| "eval_loss": 0.5452161431312561, | |
| "eval_runtime": 64.3054, | |
| "eval_samples_per_second": 9.47, | |
| "eval_steps_per_second": 0.606, | |
| "step": 207 | |
| }, | |
| { | |
| "epoch": 9.133333333333333, | |
| "grad_norm": 24.97603416442871, | |
| "learning_rate": 2.5252525252525253e-06, | |
| "loss": 0.1358, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 9.577777777777778, | |
| "grad_norm": 27.814254760742188, | |
| "learning_rate": 0.0, | |
| "loss": 0.1415, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 9.577777777777778, | |
| "eval_accuracy": 0.8210180623973727, | |
| "eval_loss": 0.5352054834365845, | |
| "eval_runtime": 66.8384, | |
| "eval_samples_per_second": 9.112, | |
| "eval_steps_per_second": 0.583, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 9.577777777777778, | |
| "step": 220, | |
| "total_flos": 3.4259009619205325e+17, | |
| "train_loss": 0.2875923281366175, | |
| "train_runtime": 2347.1095, | |
| "train_samples_per_second": 6.131, | |
| "train_steps_per_second": 0.094 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 220, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 10, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.4259009619205325e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |