| { | |
| "best_metric": 0.8709677419354839, | |
| "best_model_checkpoint": "beit-base-patch16-224-OT-3\\checkpoint-187", | |
| "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": 0.5, | |
| "eval_loss": 1.3787554502487183, | |
| "eval_runtime": 2.4212, | |
| "eval_samples_per_second": 25.607, | |
| "eval_steps_per_second": 1.652, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 1.82, | |
| "learning_rate": 2e-05, | |
| "loss": 1.3775, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.5161290322580645, | |
| "eval_loss": 1.3397387266159058, | |
| "eval_runtime": 2.3555, | |
| "eval_samples_per_second": 26.322, | |
| "eval_steps_per_second": 1.698, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 2.91, | |
| "eval_accuracy": 0.5161290322580645, | |
| "eval_loss": 1.2779887914657593, | |
| "eval_runtime": 2.4185, | |
| "eval_samples_per_second": 25.636, | |
| "eval_steps_per_second": 1.654, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 3.64, | |
| "learning_rate": 4e-05, | |
| "loss": 1.2993, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.6774193548387096, | |
| "eval_loss": 1.1642482280731201, | |
| "eval_runtime": 2.4096, | |
| "eval_samples_per_second": 25.73, | |
| "eval_steps_per_second": 1.66, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 4.91, | |
| "eval_accuracy": 0.7096774193548387, | |
| "eval_loss": 1.0751746892929077, | |
| "eval_runtime": 2.4611, | |
| "eval_samples_per_second": 25.192, | |
| "eval_steps_per_second": 1.625, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 5.45, | |
| "learning_rate": 3.777777777777778e-05, | |
| "loss": 1.1713, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "eval_accuracy": 0.7258064516129032, | |
| "eval_loss": 0.9749240875244141, | |
| "eval_runtime": 2.3739, | |
| "eval_samples_per_second": 26.117, | |
| "eval_steps_per_second": 1.685, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 6.91, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.8665702939033508, | |
| "eval_runtime": 2.3934, | |
| "eval_samples_per_second": 25.905, | |
| "eval_steps_per_second": 1.671, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 7.27, | |
| "learning_rate": 3.555555555555555e-05, | |
| "loss": 0.9956, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.7634177803993225, | |
| "eval_runtime": 2.4661, | |
| "eval_samples_per_second": 25.141, | |
| "eval_steps_per_second": 1.622, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 8.91, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.6863434910774231, | |
| "eval_runtime": 2.3658, | |
| "eval_samples_per_second": 26.206, | |
| "eval_steps_per_second": 1.691, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 9.09, | |
| "learning_rate": 3.3333333333333335e-05, | |
| "loss": 0.845, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.600502610206604, | |
| "eval_runtime": 2.4295, | |
| "eval_samples_per_second": 25.52, | |
| "eval_steps_per_second": 1.646, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "learning_rate": 3.111111111111112e-05, | |
| "loss": 0.7145, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 10.91, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.536449134349823, | |
| "eval_runtime": 2.3554, | |
| "eval_samples_per_second": 26.323, | |
| "eval_steps_per_second": 1.698, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.5585461258888245, | |
| "eval_runtime": 2.4021, | |
| "eval_samples_per_second": 25.811, | |
| "eval_steps_per_second": 1.665, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 12.73, | |
| "learning_rate": 2.888888888888889e-05, | |
| "loss": 0.5907, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 12.91, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.4962176978588104, | |
| "eval_runtime": 2.4391, | |
| "eval_samples_per_second": 25.42, | |
| "eval_steps_per_second": 1.64, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.5070138573646545, | |
| "eval_runtime": 2.4161, | |
| "eval_samples_per_second": 25.661, | |
| "eval_steps_per_second": 1.656, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 14.55, | |
| "learning_rate": 2.6666666666666667e-05, | |
| "loss": 0.5531, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 14.91, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.4648016691207886, | |
| "eval_runtime": 2.7067, | |
| "eval_samples_per_second": 22.906, | |
| "eval_steps_per_second": 1.478, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.7580645161290323, | |
| "eval_loss": 0.48115310072898865, | |
| "eval_runtime": 2.4644, | |
| "eval_samples_per_second": 25.159, | |
| "eval_steps_per_second": 1.623, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 16.36, | |
| "learning_rate": 2.444444444444445e-05, | |
| "loss": 0.4878, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 16.91, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.4281374216079712, | |
| "eval_runtime": 2.5496, | |
| "eval_samples_per_second": 24.318, | |
| "eval_steps_per_second": 1.569, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "eval_accuracy": 0.7419354838709677, | |
| "eval_loss": 0.4796009659767151, | |
| "eval_runtime": 2.389, | |
| "eval_samples_per_second": 25.953, | |
| "eval_steps_per_second": 1.674, | |
| "step": 99 | |
| }, | |
| { | |
| "epoch": 18.18, | |
| "learning_rate": 2.2222222222222227e-05, | |
| "loss": 0.4583, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 18.91, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.39126381278038025, | |
| "eval_runtime": 2.3424, | |
| "eval_samples_per_second": 26.469, | |
| "eval_steps_per_second": 1.708, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.4546, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.40853404998779297, | |
| "eval_runtime": 2.3309, | |
| "eval_samples_per_second": 26.599, | |
| "eval_steps_per_second": 1.716, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 20.91, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.4015849232673645, | |
| "eval_runtime": 2.354, | |
| "eval_samples_per_second": 26.339, | |
| "eval_steps_per_second": 1.699, | |
| "step": 115 | |
| }, | |
| { | |
| "epoch": 21.82, | |
| "learning_rate": 1.7777777777777777e-05, | |
| "loss": 0.4118, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 22.0, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.4125096797943115, | |
| "eval_runtime": 2.4266, | |
| "eval_samples_per_second": 25.55, | |
| "eval_steps_per_second": 1.648, | |
| "step": 121 | |
| }, | |
| { | |
| "epoch": 22.91, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.4282244145870209, | |
| "eval_runtime": 2.4479, | |
| "eval_samples_per_second": 25.328, | |
| "eval_steps_per_second": 1.634, | |
| "step": 126 | |
| }, | |
| { | |
| "epoch": 23.64, | |
| "learning_rate": 1.555555555555556e-05, | |
| "loss": 0.3939, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.7741935483870968, | |
| "eval_loss": 0.4869263470172882, | |
| "eval_runtime": 2.3895, | |
| "eval_samples_per_second": 25.946, | |
| "eval_steps_per_second": 1.674, | |
| "step": 132 | |
| }, | |
| { | |
| "epoch": 24.91, | |
| "eval_accuracy": 0.8548387096774194, | |
| "eval_loss": 0.37228161096572876, | |
| "eval_runtime": 2.3991, | |
| "eval_samples_per_second": 25.843, | |
| "eval_steps_per_second": 1.667, | |
| "step": 137 | |
| }, | |
| { | |
| "epoch": 25.45, | |
| "learning_rate": 1.3333333333333333e-05, | |
| "loss": 0.4138, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 26.0, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.403173565864563, | |
| "eval_runtime": 2.407, | |
| "eval_samples_per_second": 25.759, | |
| "eval_steps_per_second": 1.662, | |
| "step": 143 | |
| }, | |
| { | |
| "epoch": 26.91, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.4396969974040985, | |
| "eval_runtime": 2.4874, | |
| "eval_samples_per_second": 24.926, | |
| "eval_steps_per_second": 1.608, | |
| "step": 148 | |
| }, | |
| { | |
| "epoch": 27.27, | |
| "learning_rate": 1.1111111111111113e-05, | |
| "loss": 0.3599, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8548387096774194, | |
| "eval_loss": 0.37135955691337585, | |
| "eval_runtime": 2.3615, | |
| "eval_samples_per_second": 26.254, | |
| "eval_steps_per_second": 1.694, | |
| "step": 154 | |
| }, | |
| { | |
| "epoch": 28.91, | |
| "eval_accuracy": 0.8548387096774194, | |
| "eval_loss": 0.37999215722084045, | |
| "eval_runtime": 2.4349, | |
| "eval_samples_per_second": 25.463, | |
| "eval_steps_per_second": 1.643, | |
| "step": 159 | |
| }, | |
| { | |
| "epoch": 29.09, | |
| "learning_rate": 8.888888888888888e-06, | |
| "loss": 0.3629, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 30.0, | |
| "eval_accuracy": 0.8064516129032258, | |
| "eval_loss": 0.41584670543670654, | |
| "eval_runtime": 2.4749, | |
| "eval_samples_per_second": 25.052, | |
| "eval_steps_per_second": 1.616, | |
| "step": 165 | |
| }, | |
| { | |
| "epoch": 30.91, | |
| "learning_rate": 6.666666666666667e-06, | |
| "loss": 0.336, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 30.91, | |
| "eval_accuracy": 0.8225806451612904, | |
| "eval_loss": 0.4099889099597931, | |
| "eval_runtime": 2.447, | |
| "eval_samples_per_second": 25.337, | |
| "eval_steps_per_second": 1.635, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.8387096774193549, | |
| "eval_loss": 0.4001285433769226, | |
| "eval_runtime": 2.3315, | |
| "eval_samples_per_second": 26.592, | |
| "eval_steps_per_second": 1.716, | |
| "step": 176 | |
| }, | |
| { | |
| "epoch": 32.73, | |
| "learning_rate": 4.444444444444444e-06, | |
| "loss": 0.3306, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 32.91, | |
| "eval_accuracy": 0.8548387096774194, | |
| "eval_loss": 0.3925008773803711, | |
| "eval_runtime": 2.5275, | |
| "eval_samples_per_second": 24.53, | |
| "eval_steps_per_second": 1.583, | |
| "step": 181 | |
| }, | |
| { | |
| "epoch": 34.0, | |
| "eval_accuracy": 0.8709677419354839, | |
| "eval_loss": 0.39320024847984314, | |
| "eval_runtime": 2.343, | |
| "eval_samples_per_second": 26.462, | |
| "eval_steps_per_second": 1.707, | |
| "step": 187 | |
| }, | |
| { | |
| "epoch": 34.55, | |
| "learning_rate": 2.222222222222222e-06, | |
| "loss": 0.3319, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 34.91, | |
| "eval_accuracy": 0.8709677419354839, | |
| "eval_loss": 0.39424026012420654, | |
| "eval_runtime": 2.4449, | |
| "eval_samples_per_second": 25.359, | |
| "eval_steps_per_second": 1.636, | |
| "step": 192 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8709677419354839, | |
| "eval_loss": 0.3883473575115204, | |
| "eval_runtime": 2.3169, | |
| "eval_samples_per_second": 26.76, | |
| "eval_steps_per_second": 1.726, | |
| "step": 198 | |
| }, | |
| { | |
| "epoch": 36.36, | |
| "learning_rate": 0.0, | |
| "loss": 0.3324, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 36.36, | |
| "eval_accuracy": 0.8709677419354839, | |
| "eval_loss": 0.38859400153160095, | |
| "eval_runtime": 2.2999, | |
| "eval_samples_per_second": 26.958, | |
| "eval_steps_per_second": 1.739, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 36.36, | |
| "step": 200, | |
| "total_flos": 9.859141601338982e+17, | |
| "train_loss": 0.6110415649414063, | |
| "train_runtime": 726.155, | |
| "train_samples_per_second": 19.28, | |
| "train_steps_per_second": 0.275 | |
| } | |
| ], | |
| "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 | |
| } | |