| { | |
| "best_metric": 0.8653846153846154, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-phones/checkpoint-135", | |
| "epoch": 28.0, | |
| "eval_steps": 500, | |
| "global_step": 210, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9333333333333333, | |
| "eval_accuracy": 0.5673076923076923, | |
| "eval_loss": 0.6742563247680664, | |
| "eval_runtime": 1.0496, | |
| "eval_samples_per_second": 99.088, | |
| "eval_steps_per_second": 3.811, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.3333333333333333, | |
| "grad_norm": 5.86643123626709, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.6763, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.6923076923076923, | |
| "eval_loss": 0.6165803670883179, | |
| "eval_runtime": 0.6494, | |
| "eval_samples_per_second": 160.146, | |
| "eval_steps_per_second": 6.159, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.6666666666666665, | |
| "grad_norm": 6.109733581542969, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.635, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.9333333333333336, | |
| "eval_accuracy": 0.7403846153846154, | |
| "eval_loss": 0.5646082758903503, | |
| "eval_runtime": 0.7406, | |
| "eval_samples_per_second": 140.428, | |
| "eval_steps_per_second": 5.401, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 14.055779457092285, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.5724, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7307692307692307, | |
| "eval_loss": 0.5073935985565186, | |
| "eval_runtime": 0.6701, | |
| "eval_samples_per_second": 155.19, | |
| "eval_steps_per_second": 5.969, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.933333333333334, | |
| "eval_accuracy": 0.7692307692307693, | |
| "eval_loss": 0.48087915778160095, | |
| "eval_runtime": 0.6729, | |
| "eval_samples_per_second": 154.555, | |
| "eval_steps_per_second": 5.944, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 5.333333333333333, | |
| "grad_norm": 8.836523056030273, | |
| "learning_rate": 4.4973544973544974e-05, | |
| "loss": 0.527, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7692307692307693, | |
| "eval_loss": 0.45965665578842163, | |
| "eval_runtime": 0.6926, | |
| "eval_samples_per_second": 150.153, | |
| "eval_steps_per_second": 5.775, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 6.666666666666667, | |
| "grad_norm": 6.653749465942383, | |
| "learning_rate": 4.232804232804233e-05, | |
| "loss": 0.5304, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.933333333333334, | |
| "eval_accuracy": 0.7596153846153846, | |
| "eval_loss": 0.47583022713661194, | |
| "eval_runtime": 0.681, | |
| "eval_samples_per_second": 152.716, | |
| "eval_steps_per_second": 5.874, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "grad_norm": 13.230646133422852, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.4597, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7884615384615384, | |
| "eval_loss": 0.43429186940193176, | |
| "eval_runtime": 0.7692, | |
| "eval_samples_per_second": 135.213, | |
| "eval_steps_per_second": 5.2, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.933333333333334, | |
| "eval_accuracy": 0.7980769230769231, | |
| "eval_loss": 0.42488300800323486, | |
| "eval_runtime": 0.6816, | |
| "eval_samples_per_second": 152.578, | |
| "eval_steps_per_second": 5.868, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 9.333333333333334, | |
| "grad_norm": 8.930418968200684, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.4606, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.7980769230769231, | |
| "eval_loss": 0.42358094453811646, | |
| "eval_runtime": 0.6708, | |
| "eval_samples_per_second": 155.044, | |
| "eval_steps_per_second": 5.963, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 10.666666666666666, | |
| "grad_norm": 10.609027862548828, | |
| "learning_rate": 3.439153439153439e-05, | |
| "loss": 0.4286, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.933333333333334, | |
| "eval_accuracy": 0.8461538461538461, | |
| "eval_loss": 0.4054819643497467, | |
| "eval_runtime": 0.6768, | |
| "eval_samples_per_second": 153.664, | |
| "eval_steps_per_second": 5.91, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "grad_norm": 5.532287120819092, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.3857, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8269230769230769, | |
| "eval_loss": 0.4144248962402344, | |
| "eval_runtime": 0.7584, | |
| "eval_samples_per_second": 137.132, | |
| "eval_steps_per_second": 5.274, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 12.933333333333334, | |
| "eval_accuracy": 0.7980769230769231, | |
| "eval_loss": 0.4293949007987976, | |
| "eval_runtime": 0.6743, | |
| "eval_samples_per_second": 154.239, | |
| "eval_steps_per_second": 5.932, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 13.333333333333334, | |
| "grad_norm": 11.760865211486816, | |
| "learning_rate": 2.91005291005291e-05, | |
| "loss": 0.3801, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.8461538461538461, | |
| "eval_loss": 0.40805691480636597, | |
| "eval_runtime": 0.6845, | |
| "eval_samples_per_second": 151.929, | |
| "eval_steps_per_second": 5.843, | |
| "step": 105 | |
| }, | |
| { | |
| "epoch": 14.666666666666666, | |
| "grad_norm": 11.491521835327148, | |
| "learning_rate": 2.6455026455026456e-05, | |
| "loss": 0.3538, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 14.933333333333334, | |
| "eval_accuracy": 0.8461538461538461, | |
| "eval_loss": 0.4194793403148651, | |
| "eval_runtime": 0.674, | |
| "eval_samples_per_second": 154.298, | |
| "eval_steps_per_second": 5.935, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "grad_norm": 11.567109107971191, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.3585, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8557692307692307, | |
| "eval_loss": 0.4068710207939148, | |
| "eval_runtime": 0.7839, | |
| "eval_samples_per_second": 132.667, | |
| "eval_steps_per_second": 5.103, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.933333333333334, | |
| "eval_accuracy": 0.8557692307692307, | |
| "eval_loss": 0.3970623016357422, | |
| "eval_runtime": 0.6669, | |
| "eval_samples_per_second": 155.948, | |
| "eval_steps_per_second": 5.998, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 17.333333333333332, | |
| "grad_norm": 12.216808319091797, | |
| "learning_rate": 2.1164021164021164e-05, | |
| "loss": 0.3258, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.8653846153846154, | |
| "eval_loss": 0.39384937286376953, | |
| "eval_runtime": 0.6899, | |
| "eval_samples_per_second": 150.747, | |
| "eval_steps_per_second": 5.798, | |
| "step": 135 | |
| }, | |
| { | |
| "epoch": 18.666666666666668, | |
| "grad_norm": 15.223637580871582, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.3288, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 18.933333333333334, | |
| "eval_accuracy": 0.8461538461538461, | |
| "eval_loss": 0.396359384059906, | |
| "eval_runtime": 0.6861, | |
| "eval_samples_per_second": 151.574, | |
| "eval_steps_per_second": 5.83, | |
| "step": 142 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "grad_norm": 25.446683883666992, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.3276, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8557692307692307, | |
| "eval_loss": 0.44233372807502747, | |
| "eval_runtime": 0.6969, | |
| "eval_samples_per_second": 149.226, | |
| "eval_steps_per_second": 5.739, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 20.933333333333334, | |
| "eval_accuracy": 0.8365384615384616, | |
| "eval_loss": 0.40670448541641235, | |
| "eval_runtime": 0.7625, | |
| "eval_samples_per_second": 136.392, | |
| "eval_steps_per_second": 5.246, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 21.333333333333332, | |
| "grad_norm": 10.679701805114746, | |
| "learning_rate": 1.3227513227513228e-05, | |
| "loss": 0.317, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8653846153846154, | |
| "eval_loss": 0.4178958535194397, | |
| "eval_runtime": 0.6933, | |
| "eval_samples_per_second": 150.017, | |
| "eval_steps_per_second": 5.77, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 22.666666666666668, | |
| "grad_norm": 12.451942443847656, | |
| "learning_rate": 1.0582010582010582e-05, | |
| "loss": 0.288, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.933333333333334, | |
| "eval_accuracy": 0.8557692307692307, | |
| "eval_loss": 0.3881677985191345, | |
| "eval_runtime": 0.6592, | |
| "eval_samples_per_second": 157.766, | |
| "eval_steps_per_second": 6.068, | |
| "step": 172 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "grad_norm": 11.666213989257812, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.2735, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8557692307692307, | |
| "eval_loss": 0.42146697640419006, | |
| "eval_runtime": 0.6742, | |
| "eval_samples_per_second": 154.246, | |
| "eval_steps_per_second": 5.933, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.933333333333334, | |
| "eval_accuracy": 0.8461538461538461, | |
| "eval_loss": 0.3971670866012573, | |
| "eval_runtime": 0.7934, | |
| "eval_samples_per_second": 131.085, | |
| "eval_steps_per_second": 5.042, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 25.333333333333332, | |
| "grad_norm": 7.033930778503418, | |
| "learning_rate": 5.291005291005291e-06, | |
| "loss": 0.2805, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8557692307692307, | |
| "eval_loss": 0.3943334221839905, | |
| "eval_runtime": 0.6986, | |
| "eval_samples_per_second": 148.874, | |
| "eval_steps_per_second": 5.726, | |
| "step": 195 | |
| }, | |
| { | |
| "epoch": 26.666666666666668, | |
| "grad_norm": 5.698335647583008, | |
| "learning_rate": 2.6455026455026455e-06, | |
| "loss": 0.2961, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 26.933333333333334, | |
| "eval_accuracy": 0.8557692307692307, | |
| "eval_loss": 0.39985355734825134, | |
| "eval_runtime": 0.691, | |
| "eval_samples_per_second": 150.514, | |
| "eval_steps_per_second": 5.789, | |
| "step": 202 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "grad_norm": 9.678221702575684, | |
| "learning_rate": 0.0, | |
| "loss": 0.2832, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8557692307692307, | |
| "eval_loss": 0.4042527377605438, | |
| "eval_runtime": 0.6919, | |
| "eval_samples_per_second": 150.307, | |
| "eval_steps_per_second": 5.781, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "step": 210, | |
| "total_flos": 6.51425759341314e+17, | |
| "train_loss": 0.4042152370725359, | |
| "train_runtime": 453.861, | |
| "train_samples_per_second": 61.869, | |
| "train_steps_per_second": 0.463 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 210, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "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": 6.51425759341314e+17, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |