| { | |
| "best_metric": 0.6333333333333333, | |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-va-U5-42B\\checkpoint-193", | |
| "epoch": 37.935483870967744, | |
| "eval_steps": 500, | |
| "global_step": 294, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 7.866283893585205, | |
| "eval_runtime": 1.0387, | |
| "eval_samples_per_second": 57.764, | |
| "eval_steps_per_second": 1.925, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 1.6000000000000003e-05, | |
| "loss": 6.936, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 7.757172584533691, | |
| "eval_runtime": 0.9676, | |
| "eval_samples_per_second": 62.01, | |
| "eval_steps_per_second": 2.067, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 7.17894983291626, | |
| "eval_runtime": 1.0595, | |
| "eval_samples_per_second": 56.633, | |
| "eval_steps_per_second": 1.888, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 6.7016, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 5.914210796356201, | |
| "eval_runtime": 1.1717, | |
| "eval_samples_per_second": 51.21, | |
| "eval_steps_per_second": 1.707, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 3.909090909090909e-05, | |
| "loss": 5.5418, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 4.606503486633301, | |
| "eval_runtime": 0.9707, | |
| "eval_samples_per_second": 61.813, | |
| "eval_steps_per_second": 2.06, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 3.2782602310180664, | |
| "eval_runtime": 1.0414, | |
| "eval_samples_per_second": 57.616, | |
| "eval_steps_per_second": 1.921, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 3.7272727272727276e-05, | |
| "loss": 3.6439, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 2.1983864307403564, | |
| "eval_runtime": 1.0791, | |
| "eval_samples_per_second": 55.604, | |
| "eval_steps_per_second": 1.853, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 3.545454545454546e-05, | |
| "loss": 2.2477, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.45, | |
| "eval_loss": 1.5630935430526733, | |
| "eval_runtime": 1.2494, | |
| "eval_samples_per_second": 48.021, | |
| "eval_steps_per_second": 1.601, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_accuracy": 0.45, | |
| "eval_loss": 1.315753698348999, | |
| "eval_runtime": 1.0403, | |
| "eval_samples_per_second": 57.675, | |
| "eval_steps_per_second": 1.923, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 9.29, | |
| "learning_rate": 3.363636363636364e-05, | |
| "loss": 1.5076, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_accuracy": 0.3333333333333333, | |
| "eval_loss": 1.3664613962173462, | |
| "eval_runtime": 1.076, | |
| "eval_samples_per_second": 55.76, | |
| "eval_steps_per_second": 1.859, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 10.84, | |
| "learning_rate": 3.181818181818182e-05, | |
| "loss": 1.3865, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.35, | |
| "eval_loss": 1.3560707569122314, | |
| "eval_runtime": 1.2112, | |
| "eval_samples_per_second": 49.538, | |
| "eval_steps_per_second": 1.651, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.4666666666666667, | |
| "eval_loss": 1.2673075199127197, | |
| "eval_runtime": 0.9395, | |
| "eval_samples_per_second": 63.867, | |
| "eval_steps_per_second": 2.129, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 12.39, | |
| "learning_rate": 3.0000000000000004e-05, | |
| "loss": 1.3436, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_accuracy": 0.2833333333333333, | |
| "eval_loss": 1.5020203590393066, | |
| "eval_runtime": 1.0828, | |
| "eval_samples_per_second": 55.411, | |
| "eval_steps_per_second": 1.847, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "learning_rate": 2.8181818181818185e-05, | |
| "loss": 1.3187, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "eval_accuracy": 0.4666666666666667, | |
| "eval_loss": 1.3018145561218262, | |
| "eval_runtime": 0.9637, | |
| "eval_samples_per_second": 62.261, | |
| "eval_steps_per_second": 2.075, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.48333333333333334, | |
| "eval_loss": 1.2581952810287476, | |
| "eval_runtime": 0.968, | |
| "eval_samples_per_second": 61.985, | |
| "eval_steps_per_second": 2.066, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 2.6363636363636365e-05, | |
| "loss": 1.2132, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.2250442504882812, | |
| "eval_runtime": 1.0265, | |
| "eval_samples_per_second": 58.453, | |
| "eval_steps_per_second": 1.948, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "eval_accuracy": 0.5166666666666667, | |
| "eval_loss": 1.2070510387420654, | |
| "eval_runtime": 0.8885, | |
| "eval_samples_per_second": 67.53, | |
| "eval_steps_per_second": 2.251, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "learning_rate": 2.454545454545455e-05, | |
| "loss": 1.2041, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "eval_accuracy": 0.5166666666666667, | |
| "eval_loss": 1.1806073188781738, | |
| "eval_runtime": 1.0077, | |
| "eval_samples_per_second": 59.543, | |
| "eval_steps_per_second": 1.985, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 18.58, | |
| "learning_rate": 2.2727272727272733e-05, | |
| "loss": 1.1756, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_accuracy": 0.55, | |
| "eval_loss": 1.1720372438430786, | |
| "eval_runtime": 1.0797, | |
| "eval_samples_per_second": 55.573, | |
| "eval_steps_per_second": 1.852, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5333333333333333, | |
| "eval_loss": 1.1318727731704712, | |
| "eval_runtime": 0.8755, | |
| "eval_samples_per_second": 68.529, | |
| "eval_steps_per_second": 2.284, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 20.13, | |
| "learning_rate": 2.090909090909091e-05, | |
| "loss": 1.1107, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0850528478622437, | |
| "eval_runtime": 0.9417, | |
| "eval_samples_per_second": 63.716, | |
| "eval_steps_per_second": 2.124, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 21.68, | |
| "learning_rate": 1.9090909090909094e-05, | |
| "loss": 1.0651, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0553667545318604, | |
| "eval_runtime": 0.9498, | |
| "eval_samples_per_second": 63.168, | |
| "eval_steps_per_second": 2.106, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.0579936504364014, | |
| "eval_runtime": 0.9211, | |
| "eval_samples_per_second": 65.141, | |
| "eval_steps_per_second": 2.171, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 1.7272727272727274e-05, | |
| "loss": 1.0419, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.1471267938613892, | |
| "eval_runtime": 0.9539, | |
| "eval_samples_per_second": 62.901, | |
| "eval_steps_per_second": 2.097, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 24.77, | |
| "learning_rate": 1.5454545454545454e-05, | |
| "loss": 0.9804, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 24.9, | |
| "eval_accuracy": 0.6333333333333333, | |
| "eval_loss": 1.0633025169372559, | |
| "eval_runtime": 0.9613, | |
| "eval_samples_per_second": 62.413, | |
| "eval_steps_per_second": 2.08, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.0597257614135742, | |
| "eval_runtime": 1.9689, | |
| "eval_samples_per_second": 30.474, | |
| "eval_steps_per_second": 1.016, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "learning_rate": 1.3636363636363637e-05, | |
| "loss": 0.9195, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "eval_accuracy": 0.6333333333333333, | |
| "eval_loss": 0.9563372135162354, | |
| "eval_runtime": 0.9494, | |
| "eval_samples_per_second": 63.199, | |
| "eval_steps_per_second": 2.107, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 27.87, | |
| "learning_rate": 1.181818181818182e-05, | |
| "loss": 0.9053, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5166666666666667, | |
| "eval_loss": 0.9838621020317078, | |
| "eval_runtime": 0.8939, | |
| "eval_samples_per_second": 67.124, | |
| "eval_steps_per_second": 2.237, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "eval_accuracy": 0.5166666666666667, | |
| "eval_loss": 0.9913681745529175, | |
| "eval_runtime": 0.8926, | |
| "eval_samples_per_second": 67.223, | |
| "eval_steps_per_second": 2.241, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 29.42, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8645, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9519664645195007, | |
| "eval_runtime": 0.9221, | |
| "eval_samples_per_second": 65.066, | |
| "eval_steps_per_second": 2.169, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 8.181818181818183e-06, | |
| "loss": 0.8139, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_accuracy": 0.5333333333333333, | |
| "eval_loss": 0.9572514295578003, | |
| "eval_runtime": 0.8817, | |
| "eval_samples_per_second": 68.051, | |
| "eval_steps_per_second": 2.268, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.5166666666666667, | |
| "eval_loss": 0.9509603977203369, | |
| "eval_runtime": 0.9232, | |
| "eval_samples_per_second": 64.991, | |
| "eval_steps_per_second": 2.166, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 32.52, | |
| "learning_rate": 6.363636363636364e-06, | |
| "loss": 0.8151, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "eval_accuracy": 0.5166666666666667, | |
| "eval_loss": 0.9468602538108826, | |
| "eval_runtime": 0.8657, | |
| "eval_samples_per_second": 69.308, | |
| "eval_steps_per_second": 2.31, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "eval_accuracy": 0.5333333333333333, | |
| "eval_loss": 0.9872100353240967, | |
| "eval_runtime": 0.9376, | |
| "eval_samples_per_second": 63.99, | |
| "eval_steps_per_second": 2.133, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 34.06, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 0.7837, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 34.97, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9278779625892639, | |
| "eval_runtime": 1.3601, | |
| "eval_samples_per_second": 44.113, | |
| "eval_steps_per_second": 1.47, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 35.61, | |
| "learning_rate": 2.7272727272727272e-06, | |
| "loss": 0.7659, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.5666666666666667, | |
| "eval_loss": 0.9173651337623596, | |
| "eval_runtime": 0.9252, | |
| "eval_samples_per_second": 64.854, | |
| "eval_steps_per_second": 2.162, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 36.9, | |
| "eval_accuracy": 0.5666666666666667, | |
| "eval_loss": 0.934675931930542, | |
| "eval_runtime": 1.2264, | |
| "eval_samples_per_second": 48.923, | |
| "eval_steps_per_second": 1.631, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 37.16, | |
| "learning_rate": 9.090909090909091e-07, | |
| "loss": 0.7835, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 37.94, | |
| "eval_accuracy": 0.5666666666666667, | |
| "eval_loss": 0.9372119903564453, | |
| "eval_runtime": 0.9199, | |
| "eval_samples_per_second": 65.224, | |
| "eval_steps_per_second": 2.174, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 37.94, | |
| "step": 294, | |
| "total_flos": 1.2027310550050406e+18, | |
| "train_loss": 1.8558632776039798, | |
| "train_runtime": 524.2652, | |
| "train_samples_per_second": 78.029, | |
| "train_steps_per_second": 0.561 | |
| } | |
| ], | |
| "logging_steps": 12, | |
| "max_steps": 294, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 42, | |
| "save_steps": 500, | |
| "total_flos": 1.2027310550050406e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |