| { | |
| "best_metric": 0.9302325581395349, | |
| "best_model_checkpoint": "vit-base-patch16-224-dmae-va-da-40B/checkpoint-107", | |
| "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": 0.92, | |
| "eval_accuracy": 0.32558139534883723, | |
| "eval_loss": 1.4669076204299927, | |
| "eval_runtime": 0.6877, | |
| "eval_samples_per_second": 62.528, | |
| "eval_steps_per_second": 2.908, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "eval_accuracy": 0.4418604651162791, | |
| "eval_loss": 1.26887047290802, | |
| "eval_runtime": 0.661, | |
| "eval_samples_per_second": 65.049, | |
| "eval_steps_per_second": 3.026, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 2.77, | |
| "eval_accuracy": 0.46511627906976744, | |
| "eval_loss": 1.1591116189956665, | |
| "eval_runtime": 0.629, | |
| "eval_samples_per_second": 68.363, | |
| "eval_steps_per_second": 3.18, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 3.08, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 1.3901, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.5813953488372093, | |
| "eval_loss": 0.977810800075531, | |
| "eval_runtime": 0.6189, | |
| "eval_samples_per_second": 69.481, | |
| "eval_steps_per_second": 3.232, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 4.92, | |
| "eval_accuracy": 0.6511627906976745, | |
| "eval_loss": 0.8885109424591064, | |
| "eval_runtime": 0.5951, | |
| "eval_samples_per_second": 72.259, | |
| "eval_steps_per_second": 3.361, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 5.85, | |
| "eval_accuracy": 0.6511627906976745, | |
| "eval_loss": 0.7884993553161621, | |
| "eval_runtime": 0.6522, | |
| "eval_samples_per_second": 65.934, | |
| "eval_steps_per_second": 3.067, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 6.15, | |
| "learning_rate": 4.62962962962963e-05, | |
| "loss": 0.9794, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 6.77, | |
| "eval_accuracy": 0.7441860465116279, | |
| "eval_loss": 0.6853659152984619, | |
| "eval_runtime": 0.675, | |
| "eval_samples_per_second": 63.705, | |
| "eval_steps_per_second": 2.963, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "eval_accuracy": 0.7674418604651163, | |
| "eval_loss": 0.5821540951728821, | |
| "eval_runtime": 0.6249, | |
| "eval_samples_per_second": 68.812, | |
| "eval_steps_per_second": 3.201, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 8.92, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.492933064699173, | |
| "eval_runtime": 0.6176, | |
| "eval_samples_per_second": 69.619, | |
| "eval_steps_per_second": 3.238, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 9.23, | |
| "learning_rate": 4.166666666666667e-05, | |
| "loss": 0.6573, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 9.85, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.4822414219379425, | |
| "eval_runtime": 0.6106, | |
| "eval_samples_per_second": 70.422, | |
| "eval_steps_per_second": 3.275, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 10.77, | |
| "eval_accuracy": 0.8372093023255814, | |
| "eval_loss": 0.45293128490448, | |
| "eval_runtime": 0.6231, | |
| "eval_samples_per_second": 69.013, | |
| "eval_steps_per_second": 3.21, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "eval_accuracy": 0.7906976744186046, | |
| "eval_loss": 0.42034050822257996, | |
| "eval_runtime": 0.6472, | |
| "eval_samples_per_second": 66.442, | |
| "eval_steps_per_second": 3.09, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 12.31, | |
| "learning_rate": 3.7037037037037037e-05, | |
| "loss": 0.4166, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 12.92, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.38893207907676697, | |
| "eval_runtime": 0.6646, | |
| "eval_samples_per_second": 64.703, | |
| "eval_steps_per_second": 3.009, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 13.85, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.36970528960227966, | |
| "eval_runtime": 0.6738, | |
| "eval_samples_per_second": 63.818, | |
| "eval_steps_per_second": 2.968, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 14.77, | |
| "eval_accuracy": 0.813953488372093, | |
| "eval_loss": 0.3991001546382904, | |
| "eval_runtime": 0.6109, | |
| "eval_samples_per_second": 70.384, | |
| "eval_steps_per_second": 3.274, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 15.38, | |
| "learning_rate": 3.240740740740741e-05, | |
| "loss": 0.3376, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.3038073778152466, | |
| "eval_runtime": 0.6212, | |
| "eval_samples_per_second": 69.217, | |
| "eval_steps_per_second": 3.219, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 16.92, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.3138751685619354, | |
| "eval_runtime": 0.6788, | |
| "eval_samples_per_second": 63.343, | |
| "eval_steps_per_second": 2.946, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 17.85, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2821277976036072, | |
| "eval_runtime": 0.6103, | |
| "eval_samples_per_second": 70.454, | |
| "eval_steps_per_second": 3.277, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 18.46, | |
| "learning_rate": 2.777777777777778e-05, | |
| "loss": 0.191, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 18.77, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2904522120952606, | |
| "eval_runtime": 0.6114, | |
| "eval_samples_per_second": 70.334, | |
| "eval_steps_per_second": 3.271, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.2616073787212372, | |
| "eval_runtime": 0.6712, | |
| "eval_samples_per_second": 64.062, | |
| "eval_steps_per_second": 2.98, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 20.92, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2636496126651764, | |
| "eval_runtime": 0.6805, | |
| "eval_samples_per_second": 63.193, | |
| "eval_steps_per_second": 2.939, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 21.54, | |
| "learning_rate": 2.314814814814815e-05, | |
| "loss": 0.2065, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 21.85, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.286418080329895, | |
| "eval_runtime": 0.611, | |
| "eval_samples_per_second": 70.371, | |
| "eval_steps_per_second": 3.273, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 22.77, | |
| "eval_accuracy": 0.8604651162790697, | |
| "eval_loss": 0.2832680940628052, | |
| "eval_runtime": 0.6085, | |
| "eval_samples_per_second": 70.668, | |
| "eval_steps_per_second": 3.287, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 24.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.2507215142250061, | |
| "eval_runtime": 0.6793, | |
| "eval_samples_per_second": 63.297, | |
| "eval_steps_per_second": 2.944, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 24.62, | |
| "learning_rate": 1.8518518518518518e-05, | |
| "loss": 0.1328, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 24.92, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2889547646045685, | |
| "eval_runtime": 0.6175, | |
| "eval_samples_per_second": 69.635, | |
| "eval_steps_per_second": 3.239, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 25.85, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.30645591020584106, | |
| "eval_runtime": 0.607, | |
| "eval_samples_per_second": 70.842, | |
| "eval_steps_per_second": 3.295, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 26.77, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.28910166025161743, | |
| "eval_runtime": 0.6138, | |
| "eval_samples_per_second": 70.052, | |
| "eval_steps_per_second": 3.258, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 27.69, | |
| "learning_rate": 1.388888888888889e-05, | |
| "loss": 0.1065, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 28.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2815239727497101, | |
| "eval_runtime": 0.7441, | |
| "eval_samples_per_second": 57.787, | |
| "eval_steps_per_second": 2.688, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 28.92, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.27533814311027527, | |
| "eval_runtime": 0.6118, | |
| "eval_samples_per_second": 70.283, | |
| "eval_steps_per_second": 3.269, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 29.85, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2767714560031891, | |
| "eval_runtime": 0.6146, | |
| "eval_samples_per_second": 69.968, | |
| "eval_steps_per_second": 3.254, | |
| "step": 97 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "learning_rate": 9.259259259259259e-06, | |
| "loss": 0.1122, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 30.77, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2863713502883911, | |
| "eval_runtime": 0.6702, | |
| "eval_samples_per_second": 64.162, | |
| "eval_steps_per_second": 2.984, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 32.0, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.25629687309265137, | |
| "eval_runtime": 0.6095, | |
| "eval_samples_per_second": 70.554, | |
| "eval_steps_per_second": 3.282, | |
| "step": 104 | |
| }, | |
| { | |
| "epoch": 32.92, | |
| "eval_accuracy": 0.9302325581395349, | |
| "eval_loss": 0.2420913577079773, | |
| "eval_runtime": 0.6092, | |
| "eval_samples_per_second": 70.587, | |
| "eval_steps_per_second": 3.283, | |
| "step": 107 | |
| }, | |
| { | |
| "epoch": 33.85, | |
| "learning_rate": 4.6296296296296296e-06, | |
| "loss": 0.0879, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 33.85, | |
| "eval_accuracy": 0.9069767441860465, | |
| "eval_loss": 0.24532586336135864, | |
| "eval_runtime": 0.613, | |
| "eval_samples_per_second": 70.15, | |
| "eval_steps_per_second": 3.263, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 34.77, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.24344570934772491, | |
| "eval_runtime": 0.6673, | |
| "eval_samples_per_second": 64.436, | |
| "eval_steps_per_second": 2.997, | |
| "step": 113 | |
| }, | |
| { | |
| "epoch": 36.0, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.24062730371952057, | |
| "eval_runtime": 0.6294, | |
| "eval_samples_per_second": 68.316, | |
| "eval_steps_per_second": 3.177, | |
| "step": 117 | |
| }, | |
| { | |
| "epoch": 36.92, | |
| "learning_rate": 0.0, | |
| "loss": 0.1082, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 36.92, | |
| "eval_accuracy": 0.8837209302325582, | |
| "eval_loss": 0.2406519055366516, | |
| "eval_runtime": 0.6219, | |
| "eval_samples_per_second": 69.138, | |
| "eval_steps_per_second": 3.216, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 36.92, | |
| "step": 120, | |
| "total_flos": 1.1038156860172861e+18, | |
| "train_loss": 0.39384828855594, | |
| "train_runtime": 762.8266, | |
| "train_samples_per_second": 20.188, | |
| "train_steps_per_second": 0.157 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 120, | |
| "num_train_epochs": 40, | |
| "save_steps": 500, | |
| "total_flos": 1.1038156860172861e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |