| { | |
| "best_metric": 0.8431372549019608, | |
| "best_model_checkpoint": "vit-base-patch16-224-ve-b-U10-40\\checkpoint-187", | |
| "epoch": 38.4, | |
| "eval_steps": 500, | |
| "global_step": 240, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.96, | |
| "eval_accuracy": 0.2549019607843137, | |
| "eval_loss": 1.3845468759536743, | |
| "eval_runtime": 0.7849, | |
| "eval_samples_per_second": 64.976, | |
| "eval_steps_per_second": 2.548, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 1.6, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 1.3817, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 1.92, | |
| "eval_accuracy": 0.47058823529411764, | |
| "eval_loss": 1.3529051542282104, | |
| "eval_runtime": 0.8463, | |
| "eval_samples_per_second": 60.263, | |
| "eval_steps_per_second": 2.363, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 2.88, | |
| "eval_accuracy": 0.5882352941176471, | |
| "eval_loss": 1.2772458791732788, | |
| "eval_runtime": 0.7833, | |
| "eval_samples_per_second": 65.107, | |
| "eval_steps_per_second": 2.553, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 3.2, | |
| "learning_rate": 5.3070175438596496e-05, | |
| "loss": 1.2986, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.39215686274509803, | |
| "eval_loss": 1.2121165990829468, | |
| "eval_runtime": 0.7995, | |
| "eval_samples_per_second": 63.788, | |
| "eval_steps_per_second": 2.502, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 4.8, | |
| "learning_rate": 5.0657894736842104e-05, | |
| "loss": 1.1298, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.96, | |
| "eval_accuracy": 0.5882352941176471, | |
| "eval_loss": 1.1163876056671143, | |
| "eval_runtime": 0.8199, | |
| "eval_samples_per_second": 62.199, | |
| "eval_steps_per_second": 2.439, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 5.92, | |
| "eval_accuracy": 0.5882352941176471, | |
| "eval_loss": 1.0879433155059814, | |
| "eval_runtime": 0.8342, | |
| "eval_samples_per_second": 61.138, | |
| "eval_steps_per_second": 2.398, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 6.4, | |
| "learning_rate": 4.824561403508772e-05, | |
| "loss": 0.9842, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 6.88, | |
| "eval_accuracy": 0.6862745098039216, | |
| "eval_loss": 0.9897979497909546, | |
| "eval_runtime": 0.8672, | |
| "eval_samples_per_second": 58.809, | |
| "eval_steps_per_second": 2.306, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 4.5833333333333334e-05, | |
| "loss": 0.8402, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7843137254901961, | |
| "eval_loss": 0.9232768416404724, | |
| "eval_runtime": 0.7992, | |
| "eval_samples_per_second": 63.811, | |
| "eval_steps_per_second": 2.502, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 8.96, | |
| "eval_accuracy": 0.6470588235294118, | |
| "eval_loss": 0.9650364518165588, | |
| "eval_runtime": 0.824, | |
| "eval_samples_per_second": 61.892, | |
| "eval_steps_per_second": 2.427, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 9.6, | |
| "learning_rate": 4.342105263157895e-05, | |
| "loss": 0.7084, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 9.92, | |
| "eval_accuracy": 0.7450980392156863, | |
| "eval_loss": 0.8243097066879272, | |
| "eval_runtime": 0.7842, | |
| "eval_samples_per_second": 65.032, | |
| "eval_steps_per_second": 2.55, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 10.88, | |
| "eval_accuracy": 0.7647058823529411, | |
| "eval_loss": 0.7988162040710449, | |
| "eval_runtime": 0.784, | |
| "eval_samples_per_second": 65.052, | |
| "eval_steps_per_second": 2.551, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 11.2, | |
| "learning_rate": 4.100877192982456e-05, | |
| "loss": 0.5914, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7450980392156863, | |
| "eval_loss": 0.811381995677948, | |
| "eval_runtime": 0.816, | |
| "eval_samples_per_second": 62.503, | |
| "eval_steps_per_second": 2.451, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 12.8, | |
| "learning_rate": 3.859649122807018e-05, | |
| "loss": 0.461, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 12.96, | |
| "eval_accuracy": 0.7450980392156863, | |
| "eval_loss": 0.7651697993278503, | |
| "eval_runtime": 0.8198, | |
| "eval_samples_per_second": 62.211, | |
| "eval_steps_per_second": 2.44, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 13.92, | |
| "eval_accuracy": 0.7450980392156863, | |
| "eval_loss": 0.7405527234077454, | |
| "eval_runtime": 0.8169, | |
| "eval_samples_per_second": 62.435, | |
| "eval_steps_per_second": 2.448, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 14.4, | |
| "learning_rate": 3.618421052631579e-05, | |
| "loss": 0.3769, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 14.88, | |
| "eval_accuracy": 0.7450980392156863, | |
| "eval_loss": 0.6915778517723083, | |
| "eval_runtime": 0.8227, | |
| "eval_samples_per_second": 61.994, | |
| "eval_steps_per_second": 2.431, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 3.377192982456141e-05, | |
| "loss": 0.3376, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7843137254901961, | |
| "eval_loss": 0.6181640028953552, | |
| "eval_runtime": 0.8744, | |
| "eval_samples_per_second": 58.327, | |
| "eval_steps_per_second": 2.287, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 16.96, | |
| "eval_accuracy": 0.6862745098039216, | |
| "eval_loss": 0.8394649028778076, | |
| "eval_runtime": 0.7768, | |
| "eval_samples_per_second": 65.656, | |
| "eval_steps_per_second": 2.575, | |
| "step": 106 | |
| }, | |
| { | |
| "epoch": 17.6, | |
| "learning_rate": 3.1359649122807015e-05, | |
| "loss": 0.2606, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 17.92, | |
| "eval_accuracy": 0.7254901960784313, | |
| "eval_loss": 0.694075882434845, | |
| "eval_runtime": 0.8331, | |
| "eval_samples_per_second": 61.214, | |
| "eval_steps_per_second": 2.401, | |
| "step": 112 | |
| }, | |
| { | |
| "epoch": 18.88, | |
| "eval_accuracy": 0.7254901960784313, | |
| "eval_loss": 0.7345377802848816, | |
| "eval_runtime": 0.7774, | |
| "eval_samples_per_second": 65.603, | |
| "eval_steps_per_second": 2.573, | |
| "step": 118 | |
| }, | |
| { | |
| "epoch": 19.2, | |
| "learning_rate": 2.894736842105263e-05, | |
| "loss": 0.2314, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7058823529411765, | |
| "eval_loss": 0.737392246723175, | |
| "eval_runtime": 0.8088, | |
| "eval_samples_per_second": 63.057, | |
| "eval_steps_per_second": 2.473, | |
| "step": 125 | |
| }, | |
| { | |
| "epoch": 20.8, | |
| "learning_rate": 2.6535087719298248e-05, | |
| "loss": 0.1907, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 20.96, | |
| "eval_accuracy": 0.7647058823529411, | |
| "eval_loss": 0.7490261793136597, | |
| "eval_runtime": 0.7924, | |
| "eval_samples_per_second": 64.36, | |
| "eval_steps_per_second": 2.524, | |
| "step": 131 | |
| }, | |
| { | |
| "epoch": 21.92, | |
| "eval_accuracy": 0.7254901960784313, | |
| "eval_loss": 0.7292450666427612, | |
| "eval_runtime": 0.8014, | |
| "eval_samples_per_second": 63.64, | |
| "eval_steps_per_second": 2.496, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 22.4, | |
| "learning_rate": 2.412280701754386e-05, | |
| "loss": 0.1804, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 22.88, | |
| "eval_accuracy": 0.7450980392156863, | |
| "eval_loss": 0.7300562262535095, | |
| "eval_runtime": 0.8, | |
| "eval_samples_per_second": 63.75, | |
| "eval_steps_per_second": 2.5, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "learning_rate": 2.1710526315789474e-05, | |
| "loss": 0.1447, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7647058823529411, | |
| "eval_loss": 0.7224156260490417, | |
| "eval_runtime": 0.8597, | |
| "eval_samples_per_second": 59.321, | |
| "eval_steps_per_second": 2.326, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 24.96, | |
| "eval_accuracy": 0.7254901960784313, | |
| "eval_loss": 0.741477906703949, | |
| "eval_runtime": 0.7938, | |
| "eval_samples_per_second": 64.246, | |
| "eval_steps_per_second": 2.519, | |
| "step": 156 | |
| }, | |
| { | |
| "epoch": 25.6, | |
| "learning_rate": 1.929824561403509e-05, | |
| "loss": 0.1537, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 25.92, | |
| "eval_accuracy": 0.7843137254901961, | |
| "eval_loss": 0.6668033003807068, | |
| "eval_runtime": 0.7863, | |
| "eval_samples_per_second": 64.862, | |
| "eval_steps_per_second": 2.544, | |
| "step": 162 | |
| }, | |
| { | |
| "epoch": 26.88, | |
| "eval_accuracy": 0.7450980392156863, | |
| "eval_loss": 0.7187540531158447, | |
| "eval_runtime": 0.7671, | |
| "eval_samples_per_second": 66.488, | |
| "eval_steps_per_second": 2.607, | |
| "step": 168 | |
| }, | |
| { | |
| "epoch": 27.2, | |
| "learning_rate": 1.6885964912280704e-05, | |
| "loss": 0.1471, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.7450980392156863, | |
| "eval_loss": 0.7290924787521362, | |
| "eval_runtime": 0.7883, | |
| "eval_samples_per_second": 64.699, | |
| "eval_steps_per_second": 2.537, | |
| "step": 175 | |
| }, | |
| { | |
| "epoch": 28.8, | |
| "learning_rate": 1.4473684210526315e-05, | |
| "loss": 0.1241, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 28.96, | |
| "eval_accuracy": 0.803921568627451, | |
| "eval_loss": 0.5918722748756409, | |
| "eval_runtime": 0.8938, | |
| "eval_samples_per_second": 57.059, | |
| "eval_steps_per_second": 2.238, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 29.92, | |
| "eval_accuracy": 0.8431372549019608, | |
| "eval_loss": 0.5210930705070496, | |
| "eval_runtime": 0.8011, | |
| "eval_samples_per_second": 63.659, | |
| "eval_steps_per_second": 2.496, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 30.4, | |
| "learning_rate": 1.206140350877193e-05, | |
| "loss": 0.1058, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 30.88, | |
| "eval_accuracy": 0.7843137254901961, | |
| "eval_loss": 0.6106600761413574, | |
| "eval_runtime": 0.7843, | |
| "eval_samples_per_second": 65.029, | |
| "eval_steps_per_second": 2.55, | |
| "step": 193 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "learning_rate": 9.649122807017545e-06, | |
| "loss": 0.1032, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.7647058823529411, | |
| "eval_loss": 0.6863448023796082, | |
| "eval_runtime": 0.7796, | |
| "eval_samples_per_second": 65.415, | |
| "eval_steps_per_second": 2.565, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 32.96, | |
| "eval_accuracy": 0.7647058823529411, | |
| "eval_loss": 0.6295260190963745, | |
| "eval_runtime": 0.7933, | |
| "eval_samples_per_second": 64.29, | |
| "eval_steps_per_second": 2.521, | |
| "step": 206 | |
| }, | |
| { | |
| "epoch": 33.6, | |
| "learning_rate": 7.2368421052631575e-06, | |
| "loss": 0.1116, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 33.92, | |
| "eval_accuracy": 0.7843137254901961, | |
| "eval_loss": 0.606120765209198, | |
| "eval_runtime": 0.8173, | |
| "eval_samples_per_second": 62.404, | |
| "eval_steps_per_second": 2.447, | |
| "step": 212 | |
| }, | |
| { | |
| "epoch": 34.88, | |
| "eval_accuracy": 0.7843137254901961, | |
| "eval_loss": 0.6609983444213867, | |
| "eval_runtime": 0.799, | |
| "eval_samples_per_second": 63.828, | |
| "eval_steps_per_second": 2.503, | |
| "step": 218 | |
| }, | |
| { | |
| "epoch": 35.2, | |
| "learning_rate": 4.824561403508772e-06, | |
| "loss": 0.0871, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.803921568627451, | |
| "eval_loss": 0.6108803749084473, | |
| "eval_runtime": 0.7961, | |
| "eval_samples_per_second": 64.062, | |
| "eval_steps_per_second": 2.512, | |
| "step": 225 | |
| }, | |
| { | |
| "epoch": 36.8, | |
| "learning_rate": 2.412280701754386e-06, | |
| "loss": 0.1037, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 36.96, | |
| "eval_accuracy": 0.7843137254901961, | |
| "eval_loss": 0.6116394996643066, | |
| "eval_runtime": 0.9045, | |
| "eval_samples_per_second": 56.385, | |
| "eval_steps_per_second": 2.211, | |
| "step": 231 | |
| }, | |
| { | |
| "epoch": 37.92, | |
| "eval_accuracy": 0.803921568627451, | |
| "eval_loss": 0.6175830960273743, | |
| "eval_runtime": 0.7933, | |
| "eval_samples_per_second": 64.287, | |
| "eval_steps_per_second": 2.521, | |
| "step": 237 | |
| }, | |
| { | |
| "epoch": 38.4, | |
| "learning_rate": 0.0, | |
| "loss": 0.0802, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 38.4, | |
| "eval_accuracy": 0.803921568627451, | |
| "eval_loss": 0.6169308423995972, | |
| "eval_runtime": 0.7955, | |
| "eval_samples_per_second": 64.107, | |
| "eval_steps_per_second": 2.514, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 38.4, | |
| "step": 240, | |
| "total_flos": 2.351149109362852e+18, | |
| "train_loss": 0.4222560256719589, | |
| "train_runtime": 519.818, | |
| "train_samples_per_second": 60.791, | |
| "train_steps_per_second": 0.462 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 240, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 2.351149109362852e+18, | |
| "train_batch_size": 32, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |