| { | |
| "best_metric": 0.8833333333333333, | |
| "best_model_checkpoint": "vit-base-patch16-224-dmae-va-U5-20-45-5e-05\\checkpoint-116", | |
| "epoch": 18.06451612903226, | |
| "eval_steps": 500, | |
| "global_step": 140, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.48333333333333334, | |
| "eval_loss": 1.360709547996521, | |
| "eval_runtime": 1.0052, | |
| "eval_samples_per_second": 59.688, | |
| "eval_steps_per_second": 1.99, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 5.37593984962406e-05, | |
| "loss": 1.3752, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.2550396919250488, | |
| "eval_runtime": 0.9591, | |
| "eval_samples_per_second": 62.556, | |
| "eval_steps_per_second": 2.085, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 4.9624060150375936e-05, | |
| "loss": 1.2809, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.65, | |
| "eval_loss": 1.1435749530792236, | |
| "eval_runtime": 0.9632, | |
| "eval_samples_per_second": 62.294, | |
| "eval_steps_per_second": 2.076, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.548872180451128e-05, | |
| "loss": 1.1273, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.038071870803833, | |
| "eval_runtime": 0.9749, | |
| "eval_samples_per_second": 61.545, | |
| "eval_steps_per_second": 2.052, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.6833333333333333, | |
| "eval_loss": 0.9619849920272827, | |
| "eval_runtime": 0.947, | |
| "eval_samples_per_second": 63.359, | |
| "eval_steps_per_second": 2.112, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.135338345864661e-05, | |
| "loss": 0.9919, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.65, | |
| "eval_loss": 0.9154461622238159, | |
| "eval_runtime": 1.0081, | |
| "eval_samples_per_second": 59.516, | |
| "eval_steps_per_second": 1.984, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 3.7218045112781954e-05, | |
| "loss": 0.8971, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.7666666666666667, | |
| "eval_loss": 0.8501840829849243, | |
| "eval_runtime": 0.9769, | |
| "eval_samples_per_second": 61.421, | |
| "eval_steps_per_second": 2.047, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 3.3082706766917295e-05, | |
| "loss": 0.8049, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.75, | |
| "eval_loss": 0.8643815517425537, | |
| "eval_runtime": 0.9995, | |
| "eval_samples_per_second": 60.028, | |
| "eval_steps_per_second": 2.001, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_accuracy": 0.7833333333333333, | |
| "eval_loss": 0.8009711503982544, | |
| "eval_runtime": 0.9482, | |
| "eval_samples_per_second": 63.276, | |
| "eval_steps_per_second": 2.109, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 2.894736842105263e-05, | |
| "loss": 0.7119, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_accuracy": 0.8333333333333334, | |
| "eval_loss": 0.7275562882423401, | |
| "eval_runtime": 0.9716, | |
| "eval_samples_per_second": 61.757, | |
| "eval_steps_per_second": 2.059, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 2.4812030075187968e-05, | |
| "loss": 0.6172, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.6698881983757019, | |
| "eval_runtime": 0.9511, | |
| "eval_samples_per_second": 63.087, | |
| "eval_steps_per_second": 2.103, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 2.0676691729323306e-05, | |
| "loss": 0.5294, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8166666666666667, | |
| "eval_loss": 0.6532084941864014, | |
| "eval_runtime": 0.9965, | |
| "eval_samples_per_second": 60.209, | |
| "eval_steps_per_second": 2.007, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 1.6541353383458648e-05, | |
| "loss": 0.4696, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 0.6264522671699524, | |
| "eval_runtime": 0.9786, | |
| "eval_samples_per_second": 61.314, | |
| "eval_steps_per_second": 2.044, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "eval_accuracy": 0.85, | |
| "eval_loss": 0.6011604070663452, | |
| "eval_runtime": 1.0534, | |
| "eval_samples_per_second": 56.959, | |
| "eval_steps_per_second": 1.899, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 1.2406015037593984e-05, | |
| "loss": 0.4074, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.8833333333333333, | |
| "eval_loss": 0.5799916982650757, | |
| "eval_runtime": 1.0374, | |
| "eval_samples_per_second": 57.836, | |
| "eval_steps_per_second": 1.928, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 8.270676691729324e-06, | |
| "loss": 0.3822, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8666666666666667, | |
| "eval_loss": 0.5691779851913452, | |
| "eval_runtime": 1.0446, | |
| "eval_samples_per_second": 57.439, | |
| "eval_steps_per_second": 1.915, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "learning_rate": 4.135338345864662e-06, | |
| "loss": 0.3651, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "eval_accuracy": 0.8, | |
| "eval_loss": 0.6064590811729431, | |
| "eval_runtime": 0.9789, | |
| "eval_samples_per_second": 61.29, | |
| "eval_steps_per_second": 2.043, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "eval_accuracy": 0.8666666666666667, | |
| "eval_loss": 0.5680865049362183, | |
| "eval_runtime": 0.9558, | |
| "eval_samples_per_second": 62.773, | |
| "eval_steps_per_second": 2.092, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "learning_rate": 0.0, | |
| "loss": 0.3731, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "eval_accuracy": 0.8666666666666667, | |
| "eval_loss": 0.5674660205841064, | |
| "eval_runtime": 1.0121, | |
| "eval_samples_per_second": 59.281, | |
| "eval_steps_per_second": 1.976, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "step": 140, | |
| "total_flos": 1.3635734913760297e+18, | |
| "train_loss": 0.738073068005698, | |
| "train_runtime": 291.1702, | |
| "train_samples_per_second": 66.902, | |
| "train_steps_per_second": 0.481 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 140, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 20, | |
| "save_steps": 500, | |
| "total_flos": 1.3635734913760297e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |