| { | |
| "best_metric": 0.9655172413793104, | |
| "best_model_checkpoint": "vit-base-patch16-224-dmae-va-da2-40/checkpoint-46", | |
| "epoch": 37.64705882352941, | |
| "eval_steps": 500, | |
| "global_step": 160, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.94, | |
| "eval_accuracy": 0.43103448275862066, | |
| "eval_loss": 1.29306960105896, | |
| "eval_runtime": 1.9553, | |
| "eval_samples_per_second": 29.663, | |
| "eval_steps_per_second": 1.023, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 1.88, | |
| "eval_accuracy": 0.5517241379310345, | |
| "eval_loss": 1.2023708820343018, | |
| "eval_runtime": 1.4877, | |
| "eval_samples_per_second": 38.987, | |
| "eval_steps_per_second": 1.344, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "learning_rate": 3.7500000000000003e-05, | |
| "loss": 1.2651, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.82, | |
| "eval_accuracy": 0.6551724137931034, | |
| "eval_loss": 0.9896121621131897, | |
| "eval_runtime": 1.3422, | |
| "eval_samples_per_second": 43.212, | |
| "eval_steps_per_second": 1.49, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7241379310344828, | |
| "eval_loss": 0.7972208261489868, | |
| "eval_runtime": 1.1676, | |
| "eval_samples_per_second": 49.674, | |
| "eval_steps_per_second": 1.713, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 4.94, | |
| "eval_accuracy": 0.6551724137931034, | |
| "eval_loss": 0.7336073517799377, | |
| "eval_runtime": 1.4618, | |
| "eval_samples_per_second": 39.678, | |
| "eval_steps_per_second": 1.368, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 5.65, | |
| "learning_rate": 4.722222222222222e-05, | |
| "loss": 0.7523, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 5.88, | |
| "eval_accuracy": 0.8103448275862069, | |
| "eval_loss": 0.5780647397041321, | |
| "eval_runtime": 1.4658, | |
| "eval_samples_per_second": 39.569, | |
| "eval_steps_per_second": 1.364, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 6.82, | |
| "eval_accuracy": 0.8793103448275862, | |
| "eval_loss": 0.4911504089832306, | |
| "eval_runtime": 1.1824, | |
| "eval_samples_per_second": 49.051, | |
| "eval_steps_per_second": 1.691, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.9137931034482759, | |
| "eval_loss": 0.41120994091033936, | |
| "eval_runtime": 1.1976, | |
| "eval_samples_per_second": 48.428, | |
| "eval_steps_per_second": 1.67, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 8.47, | |
| "learning_rate": 4.305555555555556e-05, | |
| "loss": 0.4209, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 8.94, | |
| "eval_accuracy": 0.9137931034482759, | |
| "eval_loss": 0.338276743888855, | |
| "eval_runtime": 1.1796, | |
| "eval_samples_per_second": 49.168, | |
| "eval_steps_per_second": 1.695, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 9.88, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.3129003047943115, | |
| "eval_runtime": 1.1761, | |
| "eval_samples_per_second": 49.316, | |
| "eval_steps_per_second": 1.701, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 10.82, | |
| "eval_accuracy": 0.9655172413793104, | |
| "eval_loss": 0.26601171493530273, | |
| "eval_runtime": 1.1866, | |
| "eval_samples_per_second": 48.88, | |
| "eval_steps_per_second": 1.686, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 11.29, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.2647, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.9310344827586207, | |
| "eval_loss": 0.3184232711791992, | |
| "eval_runtime": 1.1991, | |
| "eval_samples_per_second": 48.371, | |
| "eval_steps_per_second": 1.668, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 12.94, | |
| "eval_accuracy": 0.9310344827586207, | |
| "eval_loss": 0.2870759963989258, | |
| "eval_runtime": 1.1806, | |
| "eval_samples_per_second": 49.126, | |
| "eval_steps_per_second": 1.694, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 13.88, | |
| "eval_accuracy": 0.9137931034482759, | |
| "eval_loss": 0.27659285068511963, | |
| "eval_runtime": 1.4456, | |
| "eval_samples_per_second": 40.122, | |
| "eval_steps_per_second": 1.384, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 14.12, | |
| "learning_rate": 3.472222222222222e-05, | |
| "loss": 0.1743, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 14.82, | |
| "eval_accuracy": 0.896551724137931, | |
| "eval_loss": 0.2726757526397705, | |
| "eval_runtime": 1.1844, | |
| "eval_samples_per_second": 48.97, | |
| "eval_steps_per_second": 1.689, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9310344827586207, | |
| "eval_loss": 0.2282419204711914, | |
| "eval_runtime": 1.198, | |
| "eval_samples_per_second": 48.414, | |
| "eval_steps_per_second": 1.669, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "learning_rate": 3.055555555555556e-05, | |
| "loss": 0.1511, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 16.94, | |
| "eval_accuracy": 0.896551724137931, | |
| "eval_loss": 0.2891761064529419, | |
| "eval_runtime": 1.1918, | |
| "eval_samples_per_second": 48.667, | |
| "eval_steps_per_second": 1.678, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 17.88, | |
| "eval_accuracy": 0.896551724137931, | |
| "eval_loss": 0.2482166588306427, | |
| "eval_runtime": 1.1922, | |
| "eval_samples_per_second": 48.65, | |
| "eval_steps_per_second": 1.678, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 18.82, | |
| "eval_accuracy": 0.9310344827586207, | |
| "eval_loss": 0.2363000512123108, | |
| "eval_runtime": 1.4565, | |
| "eval_samples_per_second": 39.822, | |
| "eval_steps_per_second": 1.373, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 19.76, | |
| "learning_rate": 2.6388888888888892e-05, | |
| "loss": 0.1253, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.16224975883960724, | |
| "eval_runtime": 1.5442, | |
| "eval_samples_per_second": 37.56, | |
| "eval_steps_per_second": 1.295, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 20.94, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.17531706392765045, | |
| "eval_runtime": 1.1991, | |
| "eval_samples_per_second": 48.37, | |
| "eval_steps_per_second": 1.668, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 21.88, | |
| "eval_accuracy": 0.9655172413793104, | |
| "eval_loss": 0.1592673361301422, | |
| "eval_runtime": 1.186, | |
| "eval_samples_per_second": 48.905, | |
| "eval_steps_per_second": 1.686, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 22.59, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.087, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 22.82, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.1334347277879715, | |
| "eval_runtime": 1.1763, | |
| "eval_samples_per_second": 49.307, | |
| "eval_steps_per_second": 1.7, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.10879910737276077, | |
| "eval_runtime": 1.2016, | |
| "eval_samples_per_second": 48.269, | |
| "eval_steps_per_second": 1.664, | |
| "step": 102 | |
| }, | |
| { | |
| "epoch": 24.94, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.11301546543836594, | |
| "eval_runtime": 1.2067, | |
| "eval_samples_per_second": 48.064, | |
| "eval_steps_per_second": 1.657, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 25.41, | |
| "learning_rate": 1.8055555555555555e-05, | |
| "loss": 0.0856, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 25.88, | |
| "eval_accuracy": 0.9137931034482759, | |
| "eval_loss": 0.1459188610315323, | |
| "eval_runtime": 1.1981, | |
| "eval_samples_per_second": 48.408, | |
| "eval_steps_per_second": 1.669, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 26.82, | |
| "eval_accuracy": 0.9655172413793104, | |
| "eval_loss": 0.14451515674591064, | |
| "eval_runtime": 1.1942, | |
| "eval_samples_per_second": 48.568, | |
| "eval_steps_per_second": 1.675, | |
| "step": 114 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9655172413793104, | |
| "eval_loss": 0.1233600303530693, | |
| "eval_runtime": 1.1858, | |
| "eval_samples_per_second": 48.911, | |
| "eval_steps_per_second": 1.687, | |
| "step": 119 | |
| }, | |
| { | |
| "epoch": 28.24, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.081, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 28.94, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.122359499335289, | |
| "eval_runtime": 1.1811, | |
| "eval_samples_per_second": 49.106, | |
| "eval_steps_per_second": 1.693, | |
| "step": 123 | |
| }, | |
| { | |
| "epoch": 29.88, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.13033324480056763, | |
| "eval_runtime": 1.2613, | |
| "eval_samples_per_second": 45.985, | |
| "eval_steps_per_second": 1.586, | |
| "step": 127 | |
| }, | |
| { | |
| "epoch": 30.82, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.13718828558921814, | |
| "eval_runtime": 1.5095, | |
| "eval_samples_per_second": 38.423, | |
| "eval_steps_per_second": 1.325, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 31.06, | |
| "learning_rate": 9.722222222222223e-06, | |
| "loss": 0.0554, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.14210258424282074, | |
| "eval_runtime": 1.1798, | |
| "eval_samples_per_second": 49.159, | |
| "eval_steps_per_second": 1.695, | |
| "step": 136 | |
| }, | |
| { | |
| "epoch": 32.94, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.13068123161792755, | |
| "eval_runtime": 1.1849, | |
| "eval_samples_per_second": 48.949, | |
| "eval_steps_per_second": 1.688, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 33.88, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0783, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 33.88, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.12443631142377853, | |
| "eval_runtime": 1.19, | |
| "eval_samples_per_second": 48.739, | |
| "eval_steps_per_second": 1.681, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 34.82, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.11953343451023102, | |
| "eval_runtime": 1.2234, | |
| "eval_samples_per_second": 47.409, | |
| "eval_steps_per_second": 1.635, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.11707677692174911, | |
| "eval_runtime": 1.1823, | |
| "eval_samples_per_second": 49.059, | |
| "eval_steps_per_second": 1.692, | |
| "step": 153 | |
| }, | |
| { | |
| "epoch": 36.71, | |
| "learning_rate": 1.388888888888889e-06, | |
| "loss": 0.0646, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 36.94, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.1165328100323677, | |
| "eval_runtime": 1.1871, | |
| "eval_samples_per_second": 48.857, | |
| "eval_steps_per_second": 1.685, | |
| "step": 157 | |
| }, | |
| { | |
| "epoch": 37.65, | |
| "eval_accuracy": 0.9482758620689655, | |
| "eval_loss": 0.11630651354789734, | |
| "eval_runtime": 1.1868, | |
| "eval_samples_per_second": 48.872, | |
| "eval_steps_per_second": 1.685, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 37.65, | |
| "step": 160, | |
| "total_flos": 1.5182502752352338e+18, | |
| "train_loss": 0.2719826718792319, | |
| "train_runtime": 1073.8442, | |
| "train_samples_per_second": 19.37, | |
| "train_steps_per_second": 0.149 | |
| } | |
| ], | |
| "logging_steps": 12, | |
| "max_steps": 160, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 1.5182502752352338e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |