| { | |
| "best_metric": 0.6666666666666666, | |
| "best_model_checkpoint": "swinv2-tiny-patch4-window8-256-dmae-va-U5-42B\\checkpoint-209", | |
| "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": 0.8861, | |
| "eval_samples_per_second": 67.712, | |
| "eval_steps_per_second": 2.257, | |
| "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": 0.8501, | |
| "eval_samples_per_second": 70.577, | |
| "eval_steps_per_second": 2.353, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 7.178952217102051, | |
| "eval_runtime": 0.9919, | |
| "eval_samples_per_second": 60.493, | |
| "eval_steps_per_second": 2.016, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 3.2000000000000005e-05, | |
| "loss": 6.7016, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 5.903270244598389, | |
| "eval_runtime": 0.9119, | |
| "eval_samples_per_second": 65.796, | |
| "eval_steps_per_second": 2.193, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 3.909090909090909e-05, | |
| "loss": 5.5439, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 4.611630916595459, | |
| "eval_runtime": 0.8484, | |
| "eval_samples_per_second": 70.723, | |
| "eval_steps_per_second": 2.357, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 3.2829790115356445, | |
| "eval_runtime": 0.8288, | |
| "eval_samples_per_second": 72.396, | |
| "eval_steps_per_second": 2.413, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 3.7272727272727276e-05, | |
| "loss": 3.6477, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.11666666666666667, | |
| "eval_loss": 2.2014169692993164, | |
| "eval_runtime": 0.9115, | |
| "eval_samples_per_second": 65.826, | |
| "eval_steps_per_second": 2.194, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 3.545454545454546e-05, | |
| "loss": 2.2506, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.45, | |
| "eval_loss": 1.5647218227386475, | |
| "eval_runtime": 0.9989, | |
| "eval_samples_per_second": 60.066, | |
| "eval_steps_per_second": 2.002, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_accuracy": 0.45, | |
| "eval_loss": 1.3160017728805542, | |
| "eval_runtime": 0.8669, | |
| "eval_samples_per_second": 69.211, | |
| "eval_steps_per_second": 2.307, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 9.29, | |
| "learning_rate": 3.363636363636364e-05, | |
| "loss": 1.5088, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_accuracy": 0.3333333333333333, | |
| "eval_loss": 1.3676066398620605, | |
| "eval_runtime": 0.8692, | |
| "eval_samples_per_second": 69.028, | |
| "eval_steps_per_second": 2.301, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 10.84, | |
| "learning_rate": 3.181818181818182e-05, | |
| "loss": 1.3868, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.45, | |
| "eval_loss": 1.3390430212020874, | |
| "eval_runtime": 0.8862, | |
| "eval_samples_per_second": 67.708, | |
| "eval_steps_per_second": 2.257, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.38333333333333336, | |
| "eval_loss": 1.322303056716919, | |
| "eval_runtime": 0.8998, | |
| "eval_samples_per_second": 66.681, | |
| "eval_steps_per_second": 2.223, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 12.39, | |
| "learning_rate": 3.0000000000000004e-05, | |
| "loss": 1.351, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_accuracy": 0.45, | |
| "eval_loss": 1.3155896663665771, | |
| "eval_runtime": 0.867, | |
| "eval_samples_per_second": 69.206, | |
| "eval_steps_per_second": 2.307, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "learning_rate": 2.8181818181818185e-05, | |
| "loss": 1.3271, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "eval_accuracy": 0.48333333333333334, | |
| "eval_loss": 1.3484553098678589, | |
| "eval_runtime": 0.8698, | |
| "eval_samples_per_second": 68.985, | |
| "eval_steps_per_second": 2.299, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.48333333333333334, | |
| "eval_loss": 1.2645785808563232, | |
| "eval_runtime": 0.9492, | |
| "eval_samples_per_second": 63.21, | |
| "eval_steps_per_second": 2.107, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 2.6363636363636365e-05, | |
| "loss": 1.2322, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.48333333333333334, | |
| "eval_loss": 1.2307511568069458, | |
| "eval_runtime": 0.8623, | |
| "eval_samples_per_second": 69.584, | |
| "eval_steps_per_second": 2.319, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.215951681137085, | |
| "eval_runtime": 0.8482, | |
| "eval_samples_per_second": 70.734, | |
| "eval_steps_per_second": 2.358, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "learning_rate": 2.454545454545455e-05, | |
| "loss": 1.22, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.2015122175216675, | |
| "eval_runtime": 0.8406, | |
| "eval_samples_per_second": 71.379, | |
| "eval_steps_per_second": 2.379, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 18.58, | |
| "learning_rate": 2.2727272727272733e-05, | |
| "loss": 1.1899, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.2008181810379028, | |
| "eval_runtime": 0.9372, | |
| "eval_samples_per_second": 64.019, | |
| "eval_steps_per_second": 2.134, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.5, | |
| "eval_loss": 1.1605515480041504, | |
| "eval_runtime": 1.0194, | |
| "eval_samples_per_second": 58.858, | |
| "eval_steps_per_second": 1.962, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 20.13, | |
| "learning_rate": 2.090909090909091e-05, | |
| "loss": 1.109, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "eval_accuracy": 0.5666666666666667, | |
| "eval_loss": 1.118190050125122, | |
| "eval_runtime": 0.8587, | |
| "eval_samples_per_second": 69.877, | |
| "eval_steps_per_second": 2.329, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 21.68, | |
| "learning_rate": 1.9090909090909094e-05, | |
| "loss": 1.0603, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_accuracy": 0.5333333333333333, | |
| "eval_loss": 1.0855048894882202, | |
| "eval_runtime": 0.8678, | |
| "eval_samples_per_second": 69.144, | |
| "eval_steps_per_second": 2.305, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "eval_accuracy": 0.5666666666666667, | |
| "eval_loss": 1.0762869119644165, | |
| "eval_runtime": 0.9558, | |
| "eval_samples_per_second": 62.772, | |
| "eval_steps_per_second": 2.092, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 1.7272727272727274e-05, | |
| "loss": 1.0264, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 1.1152801513671875, | |
| "eval_runtime": 0.9634, | |
| "eval_samples_per_second": 62.28, | |
| "eval_steps_per_second": 2.076, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 24.77, | |
| "learning_rate": 1.5454545454545454e-05, | |
| "loss": 1.0086, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 24.9, | |
| "eval_accuracy": 0.65, | |
| "eval_loss": 1.0770429372787476, | |
| "eval_runtime": 0.8946, | |
| "eval_samples_per_second": 67.069, | |
| "eval_steps_per_second": 2.236, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "eval_accuracy": 0.6166666666666667, | |
| "eval_loss": 1.0040525197982788, | |
| "eval_runtime": 0.8805, | |
| "eval_samples_per_second": 68.142, | |
| "eval_steps_per_second": 2.271, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "learning_rate": 1.3636363636363637e-05, | |
| "loss": 0.9301, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "eval_accuracy": 0.6666666666666666, | |
| "eval_loss": 0.9636775851249695, | |
| "eval_runtime": 0.9061, | |
| "eval_samples_per_second": 66.221, | |
| "eval_steps_per_second": 2.207, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 27.87, | |
| "learning_rate": 1.181818181818182e-05, | |
| "loss": 0.9077, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9824321866035461, | |
| "eval_runtime": 0.8887, | |
| "eval_samples_per_second": 67.511, | |
| "eval_steps_per_second": 2.25, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.9485177397727966, | |
| "eval_runtime": 0.9068, | |
| "eval_samples_per_second": 66.168, | |
| "eval_steps_per_second": 2.206, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 29.42, | |
| "learning_rate": 1e-05, | |
| "loss": 0.8725, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "eval_accuracy": 0.6166666666666667, | |
| "eval_loss": 0.9294353723526001, | |
| "eval_runtime": 0.8792, | |
| "eval_samples_per_second": 68.246, | |
| "eval_steps_per_second": 2.275, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 8.181818181818183e-06, | |
| "loss": 0.8203, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_accuracy": 0.6166666666666667, | |
| "eval_loss": 0.9347657561302185, | |
| "eval_runtime": 0.8842, | |
| "eval_samples_per_second": 67.856, | |
| "eval_steps_per_second": 2.262, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.9295245409011841, | |
| "eval_runtime": 0.8786, | |
| "eval_samples_per_second": 68.289, | |
| "eval_steps_per_second": 2.276, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 32.52, | |
| "learning_rate": 6.363636363636364e-06, | |
| "loss": 0.8211, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.9167192578315735, | |
| "eval_runtime": 0.8974, | |
| "eval_samples_per_second": 66.863, | |
| "eval_steps_per_second": 2.229, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "eval_accuracy": 0.5833333333333334, | |
| "eval_loss": 0.9280623197555542, | |
| "eval_runtime": 0.8572, | |
| "eval_samples_per_second": 69.993, | |
| "eval_steps_per_second": 2.333, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 34.06, | |
| "learning_rate": 4.5454545454545455e-06, | |
| "loss": 0.7916, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 34.97, | |
| "eval_accuracy": 0.6333333333333333, | |
| "eval_loss": 0.880310595035553, | |
| "eval_runtime": 0.8918, | |
| "eval_samples_per_second": 67.281, | |
| "eval_steps_per_second": 2.243, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 35.61, | |
| "learning_rate": 2.7272727272727272e-06, | |
| "loss": 0.7822, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.6333333333333333, | |
| "eval_loss": 0.8785258531570435, | |
| "eval_runtime": 0.8791, | |
| "eval_samples_per_second": 68.25, | |
| "eval_steps_per_second": 2.275, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 36.9, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.8906211853027344, | |
| "eval_runtime": 0.8494, | |
| "eval_samples_per_second": 70.635, | |
| "eval_steps_per_second": 2.354, | |
| "step": 286 | |
| }, | |
| { | |
| "epoch": 37.16, | |
| "learning_rate": 9.090909090909091e-07, | |
| "loss": 0.7937, | |
| "step": 288 | |
| }, | |
| { | |
| "epoch": 37.94, | |
| "eval_accuracy": 0.6, | |
| "eval_loss": 0.889906108379364, | |
| "eval_runtime": 0.8734, | |
| "eval_samples_per_second": 68.698, | |
| "eval_steps_per_second": 2.29, | |
| "step": 294 | |
| }, | |
| { | |
| "epoch": 37.94, | |
| "step": 294, | |
| "total_flos": 1.2027310550050406e+18, | |
| "train_loss": 1.86217847324553, | |
| "train_runtime": 485.4466, | |
| "train_samples_per_second": 84.269, | |
| "train_steps_per_second": 0.606 | |
| } | |
| ], | |
| "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 | |
| } | |