| { | |
| "best_metric": 0.9908256880733946, | |
| "best_model_checkpoint": "vit-base-patch16-224-dmae-va-U/checkpoint-217", | |
| "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": 0.25688073394495414, | |
| "eval_loss": 1.4318687915802002, | |
| "eval_runtime": 1.4321, | |
| "eval_samples_per_second": 76.112, | |
| "eval_steps_per_second": 2.793, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 1.55, | |
| "learning_rate": 2.1428571428571428e-05, | |
| "loss": 1.3911, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 1.94, | |
| "eval_accuracy": 0.47706422018348627, | |
| "eval_loss": 1.213340163230896, | |
| "eval_runtime": 1.4871, | |
| "eval_samples_per_second": 73.295, | |
| "eval_steps_per_second": 2.69, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 2.97, | |
| "eval_accuracy": 0.6055045871559633, | |
| "eval_loss": 0.9487143158912659, | |
| "eval_runtime": 1.5892, | |
| "eval_samples_per_second": 68.589, | |
| "eval_steps_per_second": 2.517, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 3.1, | |
| "learning_rate": 4.2857142857142856e-05, | |
| "loss": 1.0766, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7155963302752294, | |
| "eval_loss": 0.6542130708694458, | |
| "eval_runtime": 1.5217, | |
| "eval_samples_per_second": 71.63, | |
| "eval_steps_per_second": 2.629, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 4.65, | |
| "learning_rate": 4.841269841269841e-05, | |
| "loss": 0.6974, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 4.9, | |
| "eval_accuracy": 0.8715596330275229, | |
| "eval_loss": 0.4643765389919281, | |
| "eval_runtime": 1.561, | |
| "eval_samples_per_second": 69.827, | |
| "eval_steps_per_second": 2.562, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "eval_accuracy": 0.8715596330275229, | |
| "eval_loss": 0.39188772439956665, | |
| "eval_runtime": 1.5611, | |
| "eval_samples_per_second": 69.823, | |
| "eval_steps_per_second": 2.562, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 6.19, | |
| "learning_rate": 4.603174603174603e-05, | |
| "loss": 0.421, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 6.97, | |
| "eval_accuracy": 0.8715596330275229, | |
| "eval_loss": 0.30943310260772705, | |
| "eval_runtime": 1.6488, | |
| "eval_samples_per_second": 66.109, | |
| "eval_steps_per_second": 2.426, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 7.74, | |
| "learning_rate": 4.3650793650793655e-05, | |
| "loss": 0.2513, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8990825688073395, | |
| "eval_loss": 0.2334030568599701, | |
| "eval_runtime": 1.6748, | |
| "eval_samples_per_second": 65.082, | |
| "eval_steps_per_second": 2.388, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 8.9, | |
| "eval_accuracy": 0.9174311926605505, | |
| "eval_loss": 0.19148482382297516, | |
| "eval_runtime": 1.68, | |
| "eval_samples_per_second": 64.88, | |
| "eval_steps_per_second": 2.381, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 9.29, | |
| "learning_rate": 4.126984126984127e-05, | |
| "loss": 0.1931, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 9.94, | |
| "eval_accuracy": 0.8807339449541285, | |
| "eval_loss": 0.24314457178115845, | |
| "eval_runtime": 1.7464, | |
| "eval_samples_per_second": 62.412, | |
| "eval_steps_per_second": 2.29, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 10.84, | |
| "learning_rate": 3.888888888888889e-05, | |
| "loss": 0.1757, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 10.97, | |
| "eval_accuracy": 0.944954128440367, | |
| "eval_loss": 0.16078265011310577, | |
| "eval_runtime": 1.6471, | |
| "eval_samples_per_second": 66.177, | |
| "eval_steps_per_second": 2.428, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.926605504587156, | |
| "eval_loss": 0.1423795074224472, | |
| "eval_runtime": 1.7337, | |
| "eval_samples_per_second": 62.87, | |
| "eval_steps_per_second": 2.307, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 12.39, | |
| "learning_rate": 3.650793650793651e-05, | |
| "loss": 0.1442, | |
| "step": 96 | |
| }, | |
| { | |
| "epoch": 12.9, | |
| "eval_accuracy": 0.944954128440367, | |
| "eval_loss": 0.1280096173286438, | |
| "eval_runtime": 1.6549, | |
| "eval_samples_per_second": 65.864, | |
| "eval_steps_per_second": 2.417, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "learning_rate": 3.412698412698413e-05, | |
| "loss": 0.1085, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 13.94, | |
| "eval_accuracy": 0.9541284403669725, | |
| "eval_loss": 0.10545489937067032, | |
| "eval_runtime": 1.6964, | |
| "eval_samples_per_second": 64.255, | |
| "eval_steps_per_second": 2.358, | |
| "step": 108 | |
| }, | |
| { | |
| "epoch": 14.97, | |
| "eval_accuracy": 0.9541284403669725, | |
| "eval_loss": 0.1080455482006073, | |
| "eval_runtime": 1.5479, | |
| "eval_samples_per_second": 70.417, | |
| "eval_steps_per_second": 2.584, | |
| "step": 116 | |
| }, | |
| { | |
| "epoch": 15.48, | |
| "learning_rate": 3.1746031746031745e-05, | |
| "loss": 0.1056, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.09967872500419617, | |
| "eval_runtime": 1.7093, | |
| "eval_samples_per_second": 63.77, | |
| "eval_steps_per_second": 2.34, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 16.9, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.11852575093507767, | |
| "eval_runtime": 1.5272, | |
| "eval_samples_per_second": 71.372, | |
| "eval_steps_per_second": 2.619, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 17.03, | |
| "learning_rate": 2.9365079365079366e-05, | |
| "loss": 0.0926, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 17.94, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.07731892913579941, | |
| "eval_runtime": 1.5604, | |
| "eval_samples_per_second": 69.856, | |
| "eval_steps_per_second": 2.564, | |
| "step": 139 | |
| }, | |
| { | |
| "epoch": 18.58, | |
| "learning_rate": 2.6984126984126984e-05, | |
| "loss": 0.103, | |
| "step": 144 | |
| }, | |
| { | |
| "epoch": 18.97, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.12785662710666656, | |
| "eval_runtime": 1.527, | |
| "eval_samples_per_second": 71.384, | |
| "eval_steps_per_second": 2.62, | |
| "step": 147 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.10433010756969452, | |
| "eval_runtime": 1.5327, | |
| "eval_samples_per_second": 71.115, | |
| "eval_steps_per_second": 2.61, | |
| "step": 155 | |
| }, | |
| { | |
| "epoch": 20.13, | |
| "learning_rate": 2.4603174603174602e-05, | |
| "loss": 0.0938, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 20.9, | |
| "eval_accuracy": 0.981651376146789, | |
| "eval_loss": 0.08241702616214752, | |
| "eval_runtime": 1.5631, | |
| "eval_samples_per_second": 69.735, | |
| "eval_steps_per_second": 2.559, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 21.68, | |
| "learning_rate": 2.2222222222222223e-05, | |
| "loss": 0.0891, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 21.94, | |
| "eval_accuracy": 0.9541284403669725, | |
| "eval_loss": 0.14486828446388245, | |
| "eval_runtime": 1.5236, | |
| "eval_samples_per_second": 71.54, | |
| "eval_steps_per_second": 2.625, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 22.97, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.13658782839775085, | |
| "eval_runtime": 1.7185, | |
| "eval_samples_per_second": 63.429, | |
| "eval_steps_per_second": 2.328, | |
| "step": 178 | |
| }, | |
| { | |
| "epoch": 23.23, | |
| "learning_rate": 1.984126984126984e-05, | |
| "loss": 0.0754, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9357798165137615, | |
| "eval_loss": 0.11480199545621872, | |
| "eval_runtime": 1.6552, | |
| "eval_samples_per_second": 65.853, | |
| "eval_steps_per_second": 2.417, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 24.77, | |
| "learning_rate": 1.746031746031746e-05, | |
| "loss": 0.0882, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 24.9, | |
| "eval_accuracy": 0.9357798165137615, | |
| "eval_loss": 0.19915136694908142, | |
| "eval_runtime": 1.6537, | |
| "eval_samples_per_second": 65.913, | |
| "eval_steps_per_second": 2.419, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 25.94, | |
| "eval_accuracy": 0.981651376146789, | |
| "eval_loss": 0.07425253838300705, | |
| "eval_runtime": 1.4966, | |
| "eval_samples_per_second": 72.83, | |
| "eval_steps_per_second": 2.673, | |
| "step": 201 | |
| }, | |
| { | |
| "epoch": 26.32, | |
| "learning_rate": 1.5079365079365079e-05, | |
| "loss": 0.078, | |
| "step": 204 | |
| }, | |
| { | |
| "epoch": 26.97, | |
| "eval_accuracy": 0.9724770642201835, | |
| "eval_loss": 0.06684111058712006, | |
| "eval_runtime": 1.5659, | |
| "eval_samples_per_second": 69.61, | |
| "eval_steps_per_second": 2.554, | |
| "step": 209 | |
| }, | |
| { | |
| "epoch": 27.87, | |
| "learning_rate": 1.2698412698412699e-05, | |
| "loss": 0.0666, | |
| "step": 216 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.9908256880733946, | |
| "eval_loss": 0.053366996347904205, | |
| "eval_runtime": 1.5382, | |
| "eval_samples_per_second": 70.862, | |
| "eval_steps_per_second": 2.6, | |
| "step": 217 | |
| }, | |
| { | |
| "epoch": 28.9, | |
| "eval_accuracy": 0.9908256880733946, | |
| "eval_loss": 0.049850545823574066, | |
| "eval_runtime": 1.5346, | |
| "eval_samples_per_second": 71.027, | |
| "eval_steps_per_second": 2.607, | |
| "step": 224 | |
| }, | |
| { | |
| "epoch": 29.42, | |
| "learning_rate": 1.0317460317460318e-05, | |
| "loss": 0.0514, | |
| "step": 228 | |
| }, | |
| { | |
| "epoch": 29.94, | |
| "eval_accuracy": 0.9724770642201835, | |
| "eval_loss": 0.043298669159412384, | |
| "eval_runtime": 1.5329, | |
| "eval_samples_per_second": 71.108, | |
| "eval_steps_per_second": 2.609, | |
| "step": 232 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "learning_rate": 7.936507936507936e-06, | |
| "loss": 0.062, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 30.97, | |
| "eval_accuracy": 0.963302752293578, | |
| "eval_loss": 0.08402539044618607, | |
| "eval_runtime": 1.7473, | |
| "eval_samples_per_second": 62.382, | |
| "eval_steps_per_second": 2.289, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.9724770642201835, | |
| "eval_loss": 0.05133233219385147, | |
| "eval_runtime": 1.6441, | |
| "eval_samples_per_second": 66.297, | |
| "eval_steps_per_second": 2.433, | |
| "step": 248 | |
| }, | |
| { | |
| "epoch": 32.52, | |
| "learning_rate": 5.555555555555556e-06, | |
| "loss": 0.0712, | |
| "step": 252 | |
| }, | |
| { | |
| "epoch": 32.9, | |
| "eval_accuracy": 0.981651376146789, | |
| "eval_loss": 0.04824218526482582, | |
| "eval_runtime": 1.6796, | |
| "eval_samples_per_second": 64.896, | |
| "eval_steps_per_second": 2.382, | |
| "step": 255 | |
| }, | |
| { | |
| "epoch": 33.94, | |
| "eval_accuracy": 0.981651376146789, | |
| "eval_loss": 0.055325187742710114, | |
| "eval_runtime": 1.5821, | |
| "eval_samples_per_second": 68.894, | |
| "eval_steps_per_second": 2.528, | |
| "step": 263 | |
| }, | |
| { | |
| "epoch": 34.06, | |
| "learning_rate": 3.1746031746031746e-06, | |
| "loss": 0.0703, | |
| "step": 264 | |
| }, | |
| { | |
| "epoch": 34.97, | |
| "eval_accuracy": 0.9724770642201835, | |
| "eval_loss": 0.060162752866744995, | |
| "eval_runtime": 1.5164, | |
| "eval_samples_per_second": 71.882, | |
| "eval_steps_per_second": 2.638, | |
| "step": 271 | |
| }, | |
| { | |
| "epoch": 35.61, | |
| "learning_rate": 7.936507936507937e-07, | |
| "loss": 0.0553, | |
| "step": 276 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.9724770642201835, | |
| "eval_loss": 0.05953967571258545, | |
| "eval_runtime": 1.5574, | |
| "eval_samples_per_second": 69.99, | |
| "eval_steps_per_second": 2.568, | |
| "step": 279 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "eval_accuracy": 0.9724770642201835, | |
| "eval_loss": 0.05953451991081238, | |
| "eval_runtime": 1.6854, | |
| "eval_samples_per_second": 64.671, | |
| "eval_steps_per_second": 2.373, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 36.13, | |
| "step": 280, | |
| "total_flos": 2.738306029282984e+18, | |
| "train_loss": 0.23923614642449786, | |
| "train_runtime": 1562.4778, | |
| "train_samples_per_second": 25.037, | |
| "train_steps_per_second": 0.179 | |
| } | |
| ], | |
| "logging_steps": 12, | |
| "max_steps": 280, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 2.738306029282984e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |