| { | |
| "best_metric": 0.9047619047619048, | |
| "best_model_checkpoint": "vit-msn-small-corect_dataset_lateral_flow_ivalidation/checkpoint-84", | |
| "epoch": 36.92307692307692, | |
| "eval_steps": 500, | |
| "global_step": 120, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.9230769230769231, | |
| "eval_accuracy": 0.6336996336996337, | |
| "eval_loss": 0.6349812746047974, | |
| "eval_runtime": 1.1638, | |
| "eval_samples_per_second": 234.566, | |
| "eval_steps_per_second": 4.296, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.8461538461538463, | |
| "eval_accuracy": 0.8021978021978022, | |
| "eval_loss": 0.5046992301940918, | |
| "eval_runtime": 1.119, | |
| "eval_samples_per_second": 243.967, | |
| "eval_steps_per_second": 4.468, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.769230769230769, | |
| "eval_accuracy": 0.8791208791208791, | |
| "eval_loss": 0.3700959086418152, | |
| "eval_runtime": 1.119, | |
| "eval_samples_per_second": 243.967, | |
| "eval_steps_per_second": 4.468, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 3.076923076923077, | |
| "grad_norm": 46.272850036621094, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.5485, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7435897435897436, | |
| "eval_loss": 0.5379207134246826, | |
| "eval_runtime": 1.1645, | |
| "eval_samples_per_second": 234.43, | |
| "eval_steps_per_second": 4.294, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 4.923076923076923, | |
| "eval_accuracy": 0.8937728937728938, | |
| "eval_loss": 0.27481070160865784, | |
| "eval_runtime": 1.1071, | |
| "eval_samples_per_second": 246.598, | |
| "eval_steps_per_second": 4.516, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.846153846153846, | |
| "eval_accuracy": 0.8974358974358975, | |
| "eval_loss": 0.30044275522232056, | |
| "eval_runtime": 1.1757, | |
| "eval_samples_per_second": 232.199, | |
| "eval_steps_per_second": 4.253, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 6.153846153846154, | |
| "grad_norm": 8.343905448913574, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.3335, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.769230769230769, | |
| "eval_accuracy": 0.8681318681318682, | |
| "eval_loss": 0.34923306107521057, | |
| "eval_runtime": 1.1527, | |
| "eval_samples_per_second": 236.837, | |
| "eval_steps_per_second": 4.338, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8974358974358975, | |
| "eval_loss": 0.24970969557762146, | |
| "eval_runtime": 1.1752, | |
| "eval_samples_per_second": 232.297, | |
| "eval_steps_per_second": 4.255, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 8.923076923076923, | |
| "eval_accuracy": 0.8315018315018315, | |
| "eval_loss": 0.43036821484565735, | |
| "eval_runtime": 1.1393, | |
| "eval_samples_per_second": 239.625, | |
| "eval_steps_per_second": 4.389, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 9.23076923076923, | |
| "grad_norm": 14.322465896606445, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.3087, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 9.846153846153847, | |
| "eval_accuracy": 0.8791208791208791, | |
| "eval_loss": 0.3478649854660034, | |
| "eval_runtime": 1.2364, | |
| "eval_samples_per_second": 220.797, | |
| "eval_steps_per_second": 4.044, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 10.76923076923077, | |
| "eval_accuracy": 0.8644688644688645, | |
| "eval_loss": 0.37963706254959106, | |
| "eval_runtime": 1.2083, | |
| "eval_samples_per_second": 225.946, | |
| "eval_steps_per_second": 4.138, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8351648351648352, | |
| "eval_loss": 0.4151943325996399, | |
| "eval_runtime": 1.1429, | |
| "eval_samples_per_second": 238.873, | |
| "eval_steps_per_second": 4.375, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 12.307692307692308, | |
| "grad_norm": 14.211008071899414, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.2614, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 12.923076923076923, | |
| "eval_accuracy": 0.9010989010989011, | |
| "eval_loss": 0.31987789273262024, | |
| "eval_runtime": 1.1044, | |
| "eval_samples_per_second": 247.187, | |
| "eval_steps_per_second": 4.527, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 13.846153846153847, | |
| "eval_accuracy": 0.8717948717948718, | |
| "eval_loss": 0.34341761469841003, | |
| "eval_runtime": 1.1686, | |
| "eval_samples_per_second": 233.619, | |
| "eval_steps_per_second": 4.279, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 14.76923076923077, | |
| "eval_accuracy": 0.8461538461538461, | |
| "eval_loss": 0.400124728679657, | |
| "eval_runtime": 1.2364, | |
| "eval_samples_per_second": 220.801, | |
| "eval_steps_per_second": 4.044, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 15.384615384615385, | |
| "grad_norm": 37.66127395629883, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.2471, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.8901098901098901, | |
| "eval_loss": 0.322020947933197, | |
| "eval_runtime": 1.1904, | |
| "eval_samples_per_second": 229.338, | |
| "eval_steps_per_second": 4.2, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 16.923076923076923, | |
| "eval_accuracy": 0.8717948717948718, | |
| "eval_loss": 0.35403817892074585, | |
| "eval_runtime": 1.1771, | |
| "eval_samples_per_second": 231.923, | |
| "eval_steps_per_second": 4.248, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 17.846153846153847, | |
| "eval_accuracy": 0.8534798534798534, | |
| "eval_loss": 0.401885062456131, | |
| "eval_runtime": 1.1593, | |
| "eval_samples_per_second": 235.494, | |
| "eval_steps_per_second": 4.313, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 18.46153846153846, | |
| "grad_norm": 17.38436508178711, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.2817, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 18.76923076923077, | |
| "eval_accuracy": 0.8974358974358975, | |
| "eval_loss": 0.3152279257774353, | |
| "eval_runtime": 1.1761, | |
| "eval_samples_per_second": 232.121, | |
| "eval_steps_per_second": 4.251, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8571428571428571, | |
| "eval_loss": 0.39776933193206787, | |
| "eval_runtime": 1.1622, | |
| "eval_samples_per_second": 234.895, | |
| "eval_steps_per_second": 4.302, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 20.923076923076923, | |
| "eval_accuracy": 0.8388278388278388, | |
| "eval_loss": 0.42894455790519714, | |
| "eval_runtime": 1.1297, | |
| "eval_samples_per_second": 241.661, | |
| "eval_steps_per_second": 4.426, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 21.53846153846154, | |
| "grad_norm": 5.773376941680908, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.2353, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 21.846153846153847, | |
| "eval_accuracy": 0.8974358974358975, | |
| "eval_loss": 0.31458553671836853, | |
| "eval_runtime": 1.1608, | |
| "eval_samples_per_second": 235.18, | |
| "eval_steps_per_second": 4.307, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 22.76923076923077, | |
| "eval_accuracy": 0.8864468864468864, | |
| "eval_loss": 0.3206166625022888, | |
| "eval_runtime": 1.1912, | |
| "eval_samples_per_second": 229.187, | |
| "eval_steps_per_second": 4.198, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.8827838827838828, | |
| "eval_loss": 0.3715476095676422, | |
| "eval_runtime": 1.2718, | |
| "eval_samples_per_second": 214.649, | |
| "eval_steps_per_second": 3.931, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 24.615384615384617, | |
| "grad_norm": 7.774621963500977, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.2339, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 24.923076923076923, | |
| "eval_accuracy": 0.8937728937728938, | |
| "eval_loss": 0.34460321068763733, | |
| "eval_runtime": 1.1136, | |
| "eval_samples_per_second": 245.157, | |
| "eval_steps_per_second": 4.49, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 25.846153846153847, | |
| "eval_accuracy": 0.9047619047619048, | |
| "eval_loss": 0.29301854968070984, | |
| "eval_runtime": 1.1263, | |
| "eval_samples_per_second": 242.377, | |
| "eval_steps_per_second": 4.439, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 26.76923076923077, | |
| "eval_accuracy": 0.8205128205128205, | |
| "eval_loss": 0.43494826555252075, | |
| "eval_runtime": 1.1806, | |
| "eval_samples_per_second": 231.241, | |
| "eval_steps_per_second": 4.235, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 27.692307692307693, | |
| "grad_norm": 14.50500774383545, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.2301, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8681318681318682, | |
| "eval_loss": 0.3630487322807312, | |
| "eval_runtime": 1.192, | |
| "eval_samples_per_second": 229.021, | |
| "eval_steps_per_second": 4.195, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 28.923076923076923, | |
| "eval_accuracy": 0.8644688644688645, | |
| "eval_loss": 0.366910845041275, | |
| "eval_runtime": 1.1667, | |
| "eval_samples_per_second": 234.003, | |
| "eval_steps_per_second": 4.286, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 29.846153846153847, | |
| "eval_accuracy": 0.7912087912087912, | |
| "eval_loss": 0.5037412047386169, | |
| "eval_runtime": 1.1322, | |
| "eval_samples_per_second": 241.114, | |
| "eval_steps_per_second": 4.416, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 30.76923076923077, | |
| "grad_norm": 5.89783239364624, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.2115, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 30.76923076923077, | |
| "eval_accuracy": 0.8827838827838828, | |
| "eval_loss": 0.34486547112464905, | |
| "eval_runtime": 1.1276, | |
| "eval_samples_per_second": 242.098, | |
| "eval_steps_per_second": 4.434, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.9010989010989011, | |
| "eval_loss": 0.328006774187088, | |
| "eval_runtime": 1.1424, | |
| "eval_samples_per_second": 238.979, | |
| "eval_steps_per_second": 4.377, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 32.92307692307692, | |
| "eval_accuracy": 0.8424908424908425, | |
| "eval_loss": 0.40313562750816345, | |
| "eval_runtime": 1.1332, | |
| "eval_samples_per_second": 240.909, | |
| "eval_steps_per_second": 4.412, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 33.84615384615385, | |
| "grad_norm": 13.135037422180176, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.2033, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 33.84615384615385, | |
| "eval_accuracy": 0.8534798534798534, | |
| "eval_loss": 0.3611760437488556, | |
| "eval_runtime": 1.1669, | |
| "eval_samples_per_second": 233.958, | |
| "eval_steps_per_second": 4.285, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 34.76923076923077, | |
| "eval_accuracy": 0.8901098901098901, | |
| "eval_loss": 0.31633052229881287, | |
| "eval_runtime": 1.1155, | |
| "eval_samples_per_second": 244.743, | |
| "eval_steps_per_second": 4.482, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8864468864468864, | |
| "eval_loss": 0.3233657777309418, | |
| "eval_runtime": 1.1244, | |
| "eval_samples_per_second": 242.803, | |
| "eval_steps_per_second": 4.447, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 36.92307692307692, | |
| "grad_norm": 3.5747175216674805, | |
| "learning_rate": 0.0, | |
| "loss": 0.1807, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 36.92307692307692, | |
| "eval_accuracy": 0.8791208791208791, | |
| "eval_loss": 0.33070895075798035, | |
| "eval_runtime": 1.2006, | |
| "eval_samples_per_second": 227.387, | |
| "eval_steps_per_second": 4.165, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 36.92307692307692, | |
| "step": 120, | |
| "total_flos": 5.905752132626842e+17, | |
| "train_loss": 0.27296837071577706, | |
| "train_runtime": 300.1153, | |
| "train_samples_per_second": 108.891, | |
| "train_steps_per_second": 0.4 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 120, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 5.905752132626842e+17, | |
| "train_batch_size": 64, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |