| { | |
| "best_metric": 0.6672587329780936, | |
| "best_model_checkpoint": "Distilled-Melanoma-Classification-MobileNetV2/checkpoint-60792", | |
| "epoch": 30.0, | |
| "eval_steps": 500, | |
| "global_step": 75990, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4.8333333333333334e-05, | |
| "loss": 1.9302, | |
| "step": 2533 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.551805802249852, | |
| "eval_loss": 1.6089619398117065, | |
| "eval_runtime": 454.9672, | |
| "eval_samples_per_second": 11.137, | |
| "eval_steps_per_second": 1.394, | |
| "step": 2533 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 4.666666666666667e-05, | |
| "loss": 1.6357, | |
| "step": 5066 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5964081310440102, | |
| "eval_loss": 1.3391807079315186, | |
| "eval_runtime": 461.1878, | |
| "eval_samples_per_second": 10.987, | |
| "eval_steps_per_second": 1.375, | |
| "step": 5066 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 4.5e-05, | |
| "loss": 1.5009, | |
| "step": 7599 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.6208802052496546, | |
| "eval_loss": 1.2432700395584106, | |
| "eval_runtime": 461.7485, | |
| "eval_samples_per_second": 10.974, | |
| "eval_steps_per_second": 1.373, | |
| "step": 7599 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.3333333333333334e-05, | |
| "loss": 1.4095, | |
| "step": 10132 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6246299585553582, | |
| "eval_loss": 1.3341896533966064, | |
| "eval_runtime": 455.3869, | |
| "eval_samples_per_second": 11.127, | |
| "eval_steps_per_second": 1.392, | |
| "step": 10132 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.33, | |
| "step": 12665 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.6445628577067298, | |
| "eval_loss": 1.154502511024475, | |
| "eval_runtime": 456.8814, | |
| "eval_samples_per_second": 11.09, | |
| "eval_steps_per_second": 1.388, | |
| "step": 12665 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 4e-05, | |
| "loss": 1.2499, | |
| "step": 15198 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.6329188869153345, | |
| "eval_loss": 1.3197146654129028, | |
| "eval_runtime": 470.1267, | |
| "eval_samples_per_second": 10.778, | |
| "eval_steps_per_second": 1.349, | |
| "step": 15198 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 3.8333333333333334e-05, | |
| "loss": 1.1881, | |
| "step": 17731 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "eval_accuracy": 0.6538385632524176, | |
| "eval_loss": 1.0565587282180786, | |
| "eval_runtime": 468.3787, | |
| "eval_samples_per_second": 10.818, | |
| "eval_steps_per_second": 1.354, | |
| "step": 17731 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 3.6666666666666666e-05, | |
| "loss": 1.1184, | |
| "step": 20264 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.6660746003552398, | |
| "eval_loss": 1.036229133605957, | |
| "eval_runtime": 466.4784, | |
| "eval_samples_per_second": 10.862, | |
| "eval_steps_per_second": 1.359, | |
| "step": 20264 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 3.5e-05, | |
| "loss": 1.0487, | |
| "step": 22797 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "eval_accuracy": 0.648707321886718, | |
| "eval_loss": 1.2400047779083252, | |
| "eval_runtime": 466.581, | |
| "eval_samples_per_second": 10.86, | |
| "eval_steps_per_second": 1.359, | |
| "step": 22797 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.9876, | |
| "step": 25330 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6571936056838366, | |
| "eval_loss": 1.0631685256958008, | |
| "eval_runtime": 466.0607, | |
| "eval_samples_per_second": 10.872, | |
| "eval_steps_per_second": 1.36, | |
| "step": 25330 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 3.1666666666666666e-05, | |
| "loss": 0.9002, | |
| "step": 27863 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "eval_accuracy": 0.6473258338267219, | |
| "eval_loss": 1.1159186363220215, | |
| "eval_runtime": 466.1714, | |
| "eval_samples_per_second": 10.869, | |
| "eval_steps_per_second": 1.36, | |
| "step": 27863 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.8309, | |
| "step": 30396 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6311426879810539, | |
| "eval_loss": 1.2457987070083618, | |
| "eval_runtime": 477.2863, | |
| "eval_samples_per_second": 10.616, | |
| "eval_steps_per_second": 1.328, | |
| "step": 30396 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 2.8333333333333335e-05, | |
| "loss": 0.7729, | |
| "step": 32929 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "eval_accuracy": 0.6425893033353068, | |
| "eval_loss": 1.2415727376937866, | |
| "eval_runtime": 468.9202, | |
| "eval_samples_per_second": 10.806, | |
| "eval_steps_per_second": 1.352, | |
| "step": 32929 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.7069, | |
| "step": 35462 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.6492993881981448, | |
| "eval_loss": 1.1467454433441162, | |
| "eval_runtime": 463.7522, | |
| "eval_samples_per_second": 10.926, | |
| "eval_steps_per_second": 1.367, | |
| "step": 35462 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 2.5e-05, | |
| "loss": 0.6519, | |
| "step": 37995 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "eval_accuracy": 0.6573909611209788, | |
| "eval_loss": 1.2422680854797363, | |
| "eval_runtime": 464.9266, | |
| "eval_samples_per_second": 10.898, | |
| "eval_steps_per_second": 1.364, | |
| "step": 37995 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 2.3333333333333336e-05, | |
| "loss": 0.5956, | |
| "step": 40528 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.6498914545095718, | |
| "eval_loss": 1.2838796377182007, | |
| "eval_runtime": 477.5194, | |
| "eval_samples_per_second": 10.611, | |
| "eval_steps_per_second": 1.328, | |
| "step": 40528 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 2.1666666666666667e-05, | |
| "loss": 0.5531, | |
| "step": 43061 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "eval_accuracy": 0.6591671600552595, | |
| "eval_loss": 1.228452444076538, | |
| "eval_runtime": 479.3427, | |
| "eval_samples_per_second": 10.571, | |
| "eval_steps_per_second": 1.323, | |
| "step": 43061 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.5104, | |
| "step": 45594 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.6421945924610223, | |
| "eval_loss": 1.287538766860962, | |
| "eval_runtime": 470.0202, | |
| "eval_samples_per_second": 10.78, | |
| "eval_steps_per_second": 1.349, | |
| "step": 45594 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 1.8333333333333333e-05, | |
| "loss": 0.4777, | |
| "step": 48127 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "eval_accuracy": 0.6573909611209788, | |
| "eval_loss": 1.1871315240859985, | |
| "eval_runtime": 471.7135, | |
| "eval_samples_per_second": 10.742, | |
| "eval_steps_per_second": 1.344, | |
| "step": 48127 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 1.6666666666666667e-05, | |
| "loss": 0.447, | |
| "step": 50660 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.6566015393724097, | |
| "eval_loss": 1.2442570924758911, | |
| "eval_runtime": 467.1299, | |
| "eval_samples_per_second": 10.847, | |
| "eval_steps_per_second": 1.357, | |
| "step": 50660 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "learning_rate": 1.5e-05, | |
| "loss": 0.4179, | |
| "step": 53193 | |
| }, | |
| { | |
| "epoch": 21.0, | |
| "eval_accuracy": 0.6603512926781133, | |
| "eval_loss": 1.2222002744674683, | |
| "eval_runtime": 467.1803, | |
| "eval_samples_per_second": 10.846, | |
| "eval_steps_per_second": 1.357, | |
| "step": 53193 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.3939, | |
| "step": 55726 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.6400236826524571, | |
| "eval_loss": 1.2755681276321411, | |
| "eval_runtime": 466.7893, | |
| "eval_samples_per_second": 10.855, | |
| "eval_steps_per_second": 1.358, | |
| "step": 55726 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "learning_rate": 1.1666666666666668e-05, | |
| "loss": 0.3695, | |
| "step": 58259 | |
| }, | |
| { | |
| "epoch": 23.0, | |
| "eval_accuracy": 0.656798894809552, | |
| "eval_loss": 1.1457221508026123, | |
| "eval_runtime": 468.5285, | |
| "eval_samples_per_second": 10.815, | |
| "eval_steps_per_second": 1.353, | |
| "step": 58259 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.348, | |
| "step": 60792 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.6672587329780936, | |
| "eval_loss": 1.144992709159851, | |
| "eval_runtime": 470.6423, | |
| "eval_samples_per_second": 10.766, | |
| "eval_steps_per_second": 1.347, | |
| "step": 60792 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "learning_rate": 8.333333333333334e-06, | |
| "loss": 0.3326, | |
| "step": 63325 | |
| }, | |
| { | |
| "epoch": 25.0, | |
| "eval_accuracy": 0.6644957568581015, | |
| "eval_loss": 1.0921552181243896, | |
| "eval_runtime": 470.5011, | |
| "eval_samples_per_second": 10.769, | |
| "eval_steps_per_second": 1.347, | |
| "step": 63325 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.315, | |
| "step": 65858 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.6621274916123939, | |
| "eval_loss": 1.1216638088226318, | |
| "eval_runtime": 454.9608, | |
| "eval_samples_per_second": 11.137, | |
| "eval_steps_per_second": 1.394, | |
| "step": 65858 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "learning_rate": 5e-06, | |
| "loss": 0.2995, | |
| "step": 68391 | |
| }, | |
| { | |
| "epoch": 27.0, | |
| "eval_accuracy": 0.6611407144266824, | |
| "eval_loss": 1.0976777076721191, | |
| "eval_runtime": 446.8324, | |
| "eval_samples_per_second": 11.34, | |
| "eval_steps_per_second": 1.419, | |
| "step": 68391 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "learning_rate": 3.3333333333333333e-06, | |
| "loss": 0.2943, | |
| "step": 70924 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.6607460035523979, | |
| "eval_loss": 1.1353331804275513, | |
| "eval_runtime": 452.9025, | |
| "eval_samples_per_second": 11.188, | |
| "eval_steps_per_second": 1.4, | |
| "step": 70924 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "learning_rate": 1.6666666666666667e-06, | |
| "loss": 0.2828, | |
| "step": 73457 | |
| }, | |
| { | |
| "epoch": 29.0, | |
| "eval_accuracy": 0.66350897967239, | |
| "eval_loss": 1.0929347276687622, | |
| "eval_runtime": 461.1223, | |
| "eval_samples_per_second": 10.988, | |
| "eval_steps_per_second": 1.375, | |
| "step": 73457 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.276, | |
| "step": 75990 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.6627195579238208, | |
| "eval_loss": 1.0914403200149536, | |
| "eval_runtime": 464.7038, | |
| "eval_samples_per_second": 10.904, | |
| "eval_steps_per_second": 1.364, | |
| "step": 75990 | |
| } | |
| ], | |
| "logging_steps": 500, | |
| "max_steps": 75990, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 30, | |
| "save_steps": 500, | |
| "total_flos": 1.2266522779779072e+18, | |
| "train_batch_size": 8, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |