| { | |
| "best_metric": 0.7335680751173709, | |
| "best_model_checkpoint": "vit-base-patch16-224-type/checkpoint-372", | |
| "epoch": 5.975903614457831, | |
| "eval_steps": 500, | |
| "global_step": 372, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.16, | |
| "learning_rate": 1.3157894736842106e-05, | |
| "loss": 2.2533, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.32, | |
| "learning_rate": 2.6315789473684212e-05, | |
| "loss": 2.0412, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.48, | |
| "learning_rate": 3.9473684210526316e-05, | |
| "loss": 1.8051, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.64, | |
| "learning_rate": 4.970059880239521e-05, | |
| "loss": 1.5407, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.8, | |
| "learning_rate": 4.820359281437126e-05, | |
| "loss": 1.3111, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.96, | |
| "learning_rate": 4.670658682634731e-05, | |
| "loss": 1.1891, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.6549295774647887, | |
| "eval_loss": 1.0236921310424805, | |
| "eval_runtime": 23.861, | |
| "eval_samples_per_second": 71.414, | |
| "eval_steps_per_second": 2.263, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 1.12, | |
| "learning_rate": 4.5209580838323355e-05, | |
| "loss": 1.1132, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.29, | |
| "learning_rate": 4.3712574850299406e-05, | |
| "loss": 1.0744, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.45, | |
| "learning_rate": 4.221556886227545e-05, | |
| "loss": 1.0471, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.61, | |
| "learning_rate": 4.07185628742515e-05, | |
| "loss": 1.0616, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.77, | |
| "learning_rate": 3.9221556886227544e-05, | |
| "loss": 0.9812, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.93, | |
| "learning_rate": 3.7724550898203595e-05, | |
| "loss": 0.9452, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 1.99, | |
| "eval_accuracy": 0.6948356807511737, | |
| "eval_loss": 0.8808437585830688, | |
| "eval_runtime": 23.7153, | |
| "eval_samples_per_second": 71.852, | |
| "eval_steps_per_second": 2.277, | |
| "step": 124 | |
| }, | |
| { | |
| "epoch": 2.09, | |
| "learning_rate": 3.6227544910179645e-05, | |
| "loss": 0.9356, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.25, | |
| "learning_rate": 3.473053892215569e-05, | |
| "loss": 0.906, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 2.41, | |
| "learning_rate": 3.323353293413174e-05, | |
| "loss": 0.8799, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 2.57, | |
| "learning_rate": 3.1736526946107784e-05, | |
| "loss": 0.9281, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 2.73, | |
| "learning_rate": 3.0239520958083834e-05, | |
| "loss": 0.9325, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 2.89, | |
| "learning_rate": 2.874251497005988e-05, | |
| "loss": 0.8955, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 2.99, | |
| "eval_accuracy": 0.7124413145539906, | |
| "eval_loss": 0.8164246082305908, | |
| "eval_runtime": 23.6115, | |
| "eval_samples_per_second": 72.168, | |
| "eval_steps_per_second": 2.287, | |
| "step": 186 | |
| }, | |
| { | |
| "epoch": 3.05, | |
| "learning_rate": 2.724550898203593e-05, | |
| "loss": 0.9305, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 3.21, | |
| "learning_rate": 2.5748502994011976e-05, | |
| "loss": 0.8252, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 3.37, | |
| "learning_rate": 2.4251497005988023e-05, | |
| "loss": 0.7959, | |
| "step": 210 | |
| }, | |
| { | |
| "epoch": 3.53, | |
| "learning_rate": 2.275449101796407e-05, | |
| "loss": 0.8614, | |
| "step": 220 | |
| }, | |
| { | |
| "epoch": 3.69, | |
| "learning_rate": 2.125748502994012e-05, | |
| "loss": 0.8008, | |
| "step": 230 | |
| }, | |
| { | |
| "epoch": 3.86, | |
| "learning_rate": 1.9760479041916168e-05, | |
| "loss": 0.8389, | |
| "step": 240 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.7282863849765259, | |
| "eval_loss": 0.7755224108695984, | |
| "eval_runtime": 23.4912, | |
| "eval_samples_per_second": 72.538, | |
| "eval_steps_per_second": 2.299, | |
| "step": 249 | |
| }, | |
| { | |
| "epoch": 4.02, | |
| "learning_rate": 1.8263473053892215e-05, | |
| "loss": 0.805, | |
| "step": 250 | |
| }, | |
| { | |
| "epoch": 4.18, | |
| "learning_rate": 1.6766467065868263e-05, | |
| "loss": 0.7374, | |
| "step": 260 | |
| }, | |
| { | |
| "epoch": 4.34, | |
| "learning_rate": 1.5269461077844313e-05, | |
| "loss": 0.785, | |
| "step": 270 | |
| }, | |
| { | |
| "epoch": 4.5, | |
| "learning_rate": 1.377245508982036e-05, | |
| "loss": 0.7534, | |
| "step": 280 | |
| }, | |
| { | |
| "epoch": 4.66, | |
| "learning_rate": 1.2275449101796408e-05, | |
| "loss": 0.7473, | |
| "step": 290 | |
| }, | |
| { | |
| "epoch": 4.82, | |
| "learning_rate": 1.0778443113772455e-05, | |
| "loss": 0.7854, | |
| "step": 300 | |
| }, | |
| { | |
| "epoch": 4.98, | |
| "learning_rate": 9.281437125748502e-06, | |
| "loss": 0.8038, | |
| "step": 310 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.7306338028169014, | |
| "eval_loss": 0.7651309370994568, | |
| "eval_runtime": 23.7527, | |
| "eval_samples_per_second": 71.739, | |
| "eval_steps_per_second": 2.273, | |
| "step": 311 | |
| }, | |
| { | |
| "epoch": 5.14, | |
| "learning_rate": 7.784431137724551e-06, | |
| "loss": 0.7728, | |
| "step": 320 | |
| }, | |
| { | |
| "epoch": 5.3, | |
| "learning_rate": 6.2874251497005985e-06, | |
| "loss": 0.7011, | |
| "step": 330 | |
| }, | |
| { | |
| "epoch": 5.46, | |
| "learning_rate": 4.7904191616766475e-06, | |
| "loss": 0.7337, | |
| "step": 340 | |
| }, | |
| { | |
| "epoch": 5.62, | |
| "learning_rate": 3.293413173652695e-06, | |
| "loss": 0.7479, | |
| "step": 350 | |
| }, | |
| { | |
| "epoch": 5.78, | |
| "learning_rate": 1.7964071856287426e-06, | |
| "loss": 0.7345, | |
| "step": 360 | |
| }, | |
| { | |
| "epoch": 5.94, | |
| "learning_rate": 2.9940119760479047e-07, | |
| "loss": 0.71, | |
| "step": 370 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "eval_accuracy": 0.7335680751173709, | |
| "eval_loss": 0.7623339891433716, | |
| "eval_runtime": 23.6551, | |
| "eval_samples_per_second": 72.035, | |
| "eval_steps_per_second": 2.283, | |
| "step": 372 | |
| }, | |
| { | |
| "epoch": 5.98, | |
| "step": 372, | |
| "total_flos": 3.6877651691946025e+18, | |
| "train_loss": 0.9907107369233203, | |
| "train_runtime": 1939.9637, | |
| "train_samples_per_second": 24.625, | |
| "train_steps_per_second": 0.192 | |
| } | |
| ], | |
| "logging_steps": 10, | |
| "max_steps": 372, | |
| "num_train_epochs": 6, | |
| "save_steps": 500, | |
| "total_flos": 3.6877651691946025e+18, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |