| { | |
| "best_metric": 0.9724770642201835, | |
| "best_model_checkpoint": "swin-tiny-patch4-window7-224-dmae-va-U/checkpoint-263", | |
| "epoch": 36.12903225806452, | |
| "eval_steps": 500, | |
| "global_step": 280, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9, | |
| "eval_accuracy": 0.21100917431192662, | |
| "eval_loss": 1.4643174409866333, | |
| "eval_runtime": 1.0706, | |
| "eval_samples_per_second": 101.808, | |
| "eval_steps_per_second": 3.736, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 1.4218, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.3302752293577982, | |
| "eval_loss": 1.4069517850875854, | |
| "eval_runtime": 0.7367, | |
| "eval_samples_per_second": 147.949, | |
| "eval_steps_per_second": 5.429, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.58, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 1.3226, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.3486238532110092, | |
| "eval_loss": 1.3453558683395386, | |
| "eval_runtime": 0.7271, | |
| "eval_samples_per_second": 149.91, | |
| "eval_steps_per_second": 5.501, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.87, | |
| "learning_rate": 4.960317460317461e-05, | |
| "loss": 1.1908, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.42201834862385323, | |
| "eval_loss": 1.1437890529632568, | |
| "eval_runtime": 0.7292, | |
| "eval_samples_per_second": 149.473, | |
| "eval_steps_per_second": 5.485, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.5779816513761468, | |
| "eval_loss": 0.8730040788650513, | |
| "eval_runtime": 0.8759, | |
| "eval_samples_per_second": 124.44, | |
| "eval_steps_per_second": 4.567, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.16, | |
| "learning_rate": 4.761904761904762e-05, | |
| "loss": 0.9441, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.6422018348623854, | |
| "eval_loss": 0.8099932670593262, | |
| "eval_runtime": 0.7448, | |
| "eval_samples_per_second": 146.348, | |
| "eval_steps_per_second": 5.371, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.45, | |
| "learning_rate": 4.563492063492064e-05, | |
| "loss": 0.7185, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.7339449541284404, | |
| "eval_loss": 0.6098942160606384, | |
| "eval_runtime": 0.7469, | |
| "eval_samples_per_second": 145.943, | |
| "eval_steps_per_second": 5.356, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 0.6526, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7889908256880734, | |
| "eval_loss": 0.5095982551574707, | |
| "eval_runtime": 0.8359, | |
| "eval_samples_per_second": 130.393, | |
| "eval_steps_per_second": 4.785, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_accuracy": 0.8165137614678899, | |
| "eval_loss": 0.4924605190753937, | |
| "eval_runtime": 0.7693, | |
| "eval_samples_per_second": 141.693, | |
| "eval_steps_per_second": 5.2, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 9.03, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.5185, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_accuracy": 0.8348623853211009, | |
| "eval_loss": 0.39887699484825134, | |
| "eval_runtime": 0.7535, | |
| "eval_samples_per_second": 144.667, | |
| "eval_steps_per_second": 5.309, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 10.32, | |
| "learning_rate": 3.968253968253968e-05, | |
| "loss": 0.4946, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.8807339449541285, | |
| "eval_loss": 0.3275511860847473, | |
| "eval_runtime": 0.7403, | |
| "eval_samples_per_second": 147.247, | |
| "eval_steps_per_second": 5.404, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 11.61, | |
| "learning_rate": 3.76984126984127e-05, | |
| "loss": 0.4469, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8899082568807339, | |
| "eval_loss": 0.30227306485176086, | |
| "eval_runtime": 0.8725, | |
| "eval_samples_per_second": 124.931, | |
| "eval_steps_per_second": 4.585, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "learning_rate": 3.571428571428572e-05, | |
| "loss": 0.376, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_accuracy": 0.8990825688073395, | |
| "eval_loss": 0.311217337846756, | |
| "eval_runtime": 0.7759, | |
| "eval_samples_per_second": 140.487, | |
| "eval_steps_per_second": 5.156, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "eval_accuracy": 0.926605504587156, | |
| "eval_loss": 0.21169805526733398, | |
| "eval_runtime": 0.7482, | |
| "eval_samples_per_second": 145.675, | |
| "eval_steps_per_second": 5.346, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 14.19, | |
| "learning_rate": 3.3730158730158734e-05, | |
| "loss": 0.3156, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.9174311926605505, | |
| "eval_loss": 0.202369824051857, | |
| "eval_runtime": 0.797, | |
| "eval_samples_per_second": 136.771, | |
| "eval_steps_per_second": 5.019, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.366, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.944954128440367, | |
| "eval_loss": 0.20653872191905975, | |
| "eval_runtime": 0.7725, | |
| "eval_samples_per_second": 141.108, | |
| "eval_steps_per_second": 5.178, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 16.77, | |
| "learning_rate": 2.9761904761904762e-05, | |
| "loss": 0.2806, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "eval_accuracy": 0.9174311926605505, | |
| "eval_loss": 0.19422601163387299, | |
| "eval_runtime": 0.7563, | |
| "eval_samples_per_second": 144.13, | |
| "eval_steps_per_second": 5.289, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "eval_accuracy": 0.9174311926605505, | |
| "eval_loss": 0.23925282061100006, | |
| "eval_runtime": 0.7499, | |
| "eval_samples_per_second": 145.35, | |
| "eval_steps_per_second": 5.334, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 18.06, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2695, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_accuracy": 0.9541284403669725, | |
| "eval_loss": 0.1497838944196701, | |
| "eval_runtime": 0.862, | |
| "eval_samples_per_second": 126.456, | |
| "eval_steps_per_second": 4.641, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 19.35, | |
| "learning_rate": 2.5793650793650796e-05, | |
| "loss": 0.2357, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.9357798165137615, | |
| "eval_loss": 0.14648140966892242, | |
| "eval_runtime": 0.7642, | |
| "eval_samples_per_second": 142.638, | |
| "eval_steps_per_second": 5.234, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 20.65, | |
| "learning_rate": 2.380952380952381e-05, | |
| "loss": 0.2345, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.15224653482437134, | |
| "eval_runtime": 0.7573, | |
| "eval_samples_per_second": 143.934, | |
| "eval_steps_per_second": 5.282, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "learning_rate": 2.1825396825396827e-05, | |
| "loss": 0.2157, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_accuracy": 0.944954128440367, | |
| "eval_loss": 0.14026182889938354, | |
| "eval_runtime": 0.8934, | |
| "eval_samples_per_second": 122.007, | |
| "eval_steps_per_second": 4.477, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "eval_accuracy": 0.9541284403669725, | |
| "eval_loss": 0.09988939017057419, | |
| "eval_runtime": 0.7692, | |
| "eval_samples_per_second": 141.7, | |
| "eval_steps_per_second": 5.2, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.1894, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.14273326098918915, | |
| "eval_runtime": 0.7562, | |
| "eval_samples_per_second": 144.134, | |
| "eval_steps_per_second": 5.289, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 24.52, | |
| "learning_rate": 1.785714285714286e-05, | |
| "loss": 0.2195, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 24.9, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.09490102529525757, | |
| "eval_runtime": 0.9033, | |
| "eval_samples_per_second": 120.668, | |
| "eval_steps_per_second": 4.428, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 25.81, | |
| "learning_rate": 1.5873015873015872e-05, | |
| "loss": 0.1874, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.11523288488388062, | |
| "eval_runtime": 0.7555, | |
| "eval_samples_per_second": 144.279, | |
| "eval_steps_per_second": 5.295, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "eval_accuracy": 0.9541284403669725, | |
| "eval_loss": 0.122586190700531, | |
| "eval_runtime": 0.9165, | |
| "eval_samples_per_second": 118.928, | |
| "eval_steps_per_second": 4.364, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 27.1, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.1815, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.0963682234287262, | |
| "eval_runtime": 0.7954, | |
| "eval_samples_per_second": 137.041, | |
| "eval_steps_per_second": 5.029, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 28.39, | |
| "learning_rate": 1.1904761904761905e-05, | |
| "loss": 0.1619, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.09118019044399261, | |
| "eval_runtime": 0.761, | |
| "eval_samples_per_second": 143.228, | |
| "eval_steps_per_second": 5.256, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 29.68, | |
| "learning_rate": 9.92063492063492e-06, | |
| "loss": 0.201, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.09030430763959885, | |
| "eval_runtime": 0.7673, | |
| "eval_samples_per_second": 142.058, | |
| "eval_steps_per_second": 5.213, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.1659, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.0744820311665535, | |
| "eval_runtime": 0.7579, | |
| "eval_samples_per_second": 143.812, | |
| "eval_steps_per_second": 5.278, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.07808323204517365, | |
| "eval_runtime": 0.768, | |
| "eval_samples_per_second": 141.928, | |
| "eval_steps_per_second": 5.208, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 32.26, | |
| "learning_rate": 5.9523809523809525e-06, | |
| "loss": 0.1459, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.09302664548158646, | |
| "eval_runtime": 0.763, | |
| "eval_samples_per_second": 142.857, | |
| "eval_steps_per_second": 5.242, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 33.55, | |
| "learning_rate": 3.968253968253968e-06, | |
| "loss": 0.1459, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "eval_accuracy": 0.9724770642201835, | |
| "eval_loss": 0.09002815186977386, | |
| "eval_runtime": 0.7775, | |
| "eval_samples_per_second": 140.186, | |
| "eval_steps_per_second": 5.144, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 34.84, | |
| "learning_rate": 1.984126984126984e-06, | |
| "loss": 0.1487, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 34.97, | |
| "eval_accuracy": 0.9724770642201835, | |
| "eval_loss": 0.07955732196569443, | |
| "eval_runtime": 0.8544, | |
| "eval_samples_per_second": 127.574, | |
| "eval_steps_per_second": 4.682, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9724770642201835, | |
| "eval_loss": 0.07835839688777924, | |
| "eval_runtime": 0.77, | |
| "eval_samples_per_second": 141.561, | |
| "eval_steps_per_second": 5.195, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "learning_rate": 0.0, | |
| "loss": 0.1504, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "eval_accuracy": 0.9724770642201835, | |
| "eval_loss": 0.07835046201944351, | |
| "eval_runtime": 0.8307, | |
| "eval_samples_per_second": 131.22, | |
| "eval_steps_per_second": 4.815, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "step": 280, | |
| "total_flos": 8.783603110659686e+17, | |
| "train_loss": 0.42505432409899574, | |
| "train_runtime": 631.588, | |
| "train_samples_per_second": 61.939, | |
| "train_steps_per_second": 0.443 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 280, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 8.783603110659686e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |