| { | |
| "best_metric": 0.8387096774193549, | |
| "best_model_checkpoint": "beit-base-patch16-224-OT-alt\\checkpoint-198", | |
| "epoch": 36.36363636363637, | |
| "eval_steps": 500, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.91, | |
| "eval_accuracy": 0.14516129032258066, | |
| "eval_loss": 1.7602910995483398, | |
| "eval_runtime": 2.2295, | |
| "eval_samples_per_second": 27.809, | |
| "eval_steps_per_second": 1.794, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 1.9e-05, | |
| "loss": 1.7693, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.14516129032258066, | |
| "eval_loss": 1.6916192770004272, | |
| "eval_runtime": 2.248, | |
| "eval_samples_per_second": 27.579, | |
| "eval_steps_per_second": 1.779, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_accuracy": 0.14516129032258066, | |
| "eval_loss": 1.5752112865447998, | |
| "eval_runtime": 2.2645, | |
| "eval_samples_per_second": 27.379, | |
| "eval_steps_per_second": 1.766, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 3.8e-05, | |
| "loss": 1.6261, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.14516129032258066, | |
| "eval_loss": 1.4015283584594727, | |
| "eval_runtime": 2.295, | |
| "eval_samples_per_second": 27.015, | |
| "eval_steps_per_second": 1.743, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "eval_accuracy": 0.14516129032258066, | |
| "eval_loss": 1.2889776229858398, | |
| "eval_runtime": 2.279, | |
| "eval_samples_per_second": 27.205, | |
| "eval_steps_per_second": 1.755, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 3.588888888888889e-05, | |
| "loss": 1.3534, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.3709677419354839, | |
| "eval_loss": 1.2128400802612305, | |
| "eval_runtime": 2.3157, | |
| "eval_samples_per_second": 26.774, | |
| "eval_steps_per_second": 1.727, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "eval_accuracy": 0.4032258064516129, | |
| "eval_loss": 1.1418391466140747, | |
| "eval_runtime": 2.2485, | |
| "eval_samples_per_second": 27.573, | |
| "eval_steps_per_second": 1.779, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 3.3777777777777776e-05, | |
| "loss": 1.1661, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.46774193548387094, | |
| "eval_loss": 1.0726869106292725, | |
| "eval_runtime": 2.247, | |
| "eval_samples_per_second": 27.592, | |
| "eval_steps_per_second": 1.78, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "eval_accuracy": 0.4032258064516129, | |
| "eval_loss": 1.0908629894256592, | |
| "eval_runtime": 2.288, | |
| "eval_samples_per_second": 27.097, | |
| "eval_steps_per_second": 1.748, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 3.166666666666667e-05, | |
| "loss": 1.0344, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.6129032258064516, | |
| "eval_loss": 0.9719263911247253, | |
| "eval_runtime": 2.2685, | |
| "eval_samples_per_second": 27.33, | |
| "eval_steps_per_second": 1.763, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "learning_rate": 2.9555555555555556e-05, | |
| "loss": 0.9604, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "eval_accuracy": 0.6451612903225806, | |
| "eval_loss": 0.9922870397567749, | |
| "eval_runtime": 2.249, | |
| "eval_samples_per_second": 27.567, | |
| "eval_steps_per_second": 1.779, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.6290322580645161, | |
| "eval_loss": 0.9554464221000671, | |
| "eval_runtime": 2.3255, | |
| "eval_samples_per_second": 26.661, | |
| "eval_steps_per_second": 1.72, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "learning_rate": 2.7444444444444447e-05, | |
| "loss": 0.8477, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "eval_accuracy": 0.6774193548387096, | |
| "eval_loss": 0.9155737161636353, | |
| "eval_runtime": 2.259, | |
| "eval_samples_per_second": 27.445, | |
| "eval_steps_per_second": 1.771, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7096774193548387, | |
| "eval_loss": 0.833889901638031, | |
| "eval_runtime": 2.3145, | |
| "eval_samples_per_second": 26.788, | |
| "eval_steps_per_second": 1.728, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 2.5333333333333334e-05, | |
| "loss": 0.7727, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "eval_accuracy": 0.7258064516129032, | |
| "eval_loss": 0.7850545048713684, | |
| "eval_runtime": 2.2951, | |
| "eval_samples_per_second": 27.015, | |
| "eval_steps_per_second": 1.743, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7258064516129032, | |
| "eval_loss": 0.7993963956832886, | |
| "eval_runtime": 2.2405, | |
| "eval_samples_per_second": 27.672, | |
| "eval_steps_per_second": 1.785, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 16.36, | |
| "learning_rate": 2.3222222222222224e-05, | |
| "loss": 0.6714, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 16.91, | |
| "eval_accuracy": 0.6290322580645161, | |
| "eval_loss": 0.8245651721954346, | |
| "eval_runtime": 2.1735, | |
| "eval_samples_per_second": 28.525, | |
| "eval_steps_per_second": 1.84, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7096774193548387, | |
| "eval_loss": 0.738938570022583, | |
| "eval_runtime": 2.2195, | |
| "eval_samples_per_second": 27.934, | |
| "eval_steps_per_second": 1.802, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 2.1111111111111114e-05, | |
| "loss": 0.6143, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 18.91, | |
| "eval_accuracy": 0.6451612903225806, | |
| "eval_loss": 0.8202147483825684, | |
| "eval_runtime": 2.2165, | |
| "eval_samples_per_second": 27.972, | |
| "eval_steps_per_second": 1.805, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 1.9e-05, | |
| "loss": 0.5398, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.6295467615127563, | |
| "eval_runtime": 2.241, | |
| "eval_samples_per_second": 27.666, | |
| "eval_steps_per_second": 1.785, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 20.91, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.6735957264900208, | |
| "eval_runtime": 2.3026, | |
| "eval_samples_per_second": 26.926, | |
| "eval_steps_per_second": 1.737, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 21.82, | |
| "learning_rate": 1.6888888888888888e-05, | |
| "loss": 0.4958, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.7903225806451613, | |
| "eval_loss": 0.6217902302742004, | |
| "eval_runtime": 2.2946, | |
| "eval_samples_per_second": 27.02, | |
| "eval_steps_per_second": 1.743, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 22.91, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.6400670409202576, | |
| "eval_runtime": 2.217, | |
| "eval_samples_per_second": 27.965, | |
| "eval_steps_per_second": 1.804, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 23.64, | |
| "learning_rate": 1.4777777777777778e-05, | |
| "loss": 0.4561, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7258064516129032, | |
| "eval_loss": 0.6640270352363586, | |
| "eval_runtime": 2.199, | |
| "eval_samples_per_second": 28.194, | |
| "eval_steps_per_second": 1.819, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 24.91, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.6009158492088318, | |
| "eval_runtime": 2.2165, | |
| "eval_samples_per_second": 27.972, | |
| "eval_steps_per_second": 1.805, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 25.45, | |
| "learning_rate": 1.2666666666666667e-05, | |
| "loss": 0.4149, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.5619057416915894, | |
| "eval_runtime": 2.2799, | |
| "eval_samples_per_second": 27.194, | |
| "eval_steps_per_second": 1.754, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 26.91, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.5118472576141357, | |
| "eval_runtime": 2.1675, | |
| "eval_samples_per_second": 28.604, | |
| "eval_steps_per_second": 1.845, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "learning_rate": 1.0555555555555557e-05, | |
| "loss": 0.3965, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.5682212114334106, | |
| "eval_runtime": 2.2446, | |
| "eval_samples_per_second": 27.622, | |
| "eval_steps_per_second": 1.782, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 28.91, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.5514592528343201, | |
| "eval_runtime": 2.2005, | |
| "eval_samples_per_second": 28.175, | |
| "eval_steps_per_second": 1.818, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 29.09, | |
| "learning_rate": 8.444444444444444e-06, | |
| "loss": 0.4219, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.7096774193548387, | |
| "eval_loss": 0.7044903039932251, | |
| "eval_runtime": 2.2835, | |
| "eval_samples_per_second": 27.151, | |
| "eval_steps_per_second": 1.752, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 30.91, | |
| "learning_rate": 6.333333333333333e-06, | |
| "loss": 0.3939, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 30.91, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.5711750984191895, | |
| "eval_runtime": 2.2085, | |
| "eval_samples_per_second": 28.073, | |
| "eval_steps_per_second": 1.811, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.5857284069061279, | |
| "eval_runtime": 2.3796, | |
| "eval_samples_per_second": 26.055, | |
| "eval_steps_per_second": 1.681, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 32.73, | |
| "learning_rate": 4.222222222222222e-06, | |
| "loss": 0.3598, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 32.91, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.5539482235908508, | |
| "eval_runtime": 2.3201, | |
| "eval_samples_per_second": 26.724, | |
| "eval_steps_per_second": 1.724, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.5470647215843201, | |
| "eval_runtime": 2.1645, | |
| "eval_samples_per_second": 28.644, | |
| "eval_steps_per_second": 1.848, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 34.55, | |
| "learning_rate": 2.111111111111111e-06, | |
| "loss": 0.3613, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 34.91, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.5407957434654236, | |
| "eval_runtime": 2.233, | |
| "eval_samples_per_second": 27.765, | |
| "eval_steps_per_second": 1.791, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.5388934016227722, | |
| "eval_runtime": 2.3576, | |
| "eval_samples_per_second": 26.298, | |
| "eval_steps_per_second": 1.697, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 36.36, | |
| "learning_rate": 0.0, | |
| "loss": 0.3748, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 36.36, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.538964569568634, | |
| "eval_runtime": 2.198, | |
| "eval_samples_per_second": 28.207, | |
| "eval_steps_per_second": 1.82, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 36.36, | |
| "step": 200, | |
| "total_flos": 9.859141601338982e+17, | |
| "train_loss": 0.7515219366550445, | |
| "train_runtime": 697.4947, | |
| "train_samples_per_second": 20.072, | |
| "train_steps_per_second": 0.287 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 200, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 9.859141601338982e+17, | |
| "train_batch_size": 16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |