| { | |
| "best_metric": 0.9370558857917786, | |
| "best_model_checkpoint": "./vit-base/checkpoint-200", | |
| "epoch": 20.0, | |
| "global_step": 200, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 0.00019, | |
| "loss": 0.0001, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.00018, | |
| "loss": 0.0001, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 0.00017, | |
| "loss": 0.0001, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 0.00016, | |
| "loss": 0.0001, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.00015000000000000001, | |
| "loss": 0.0001, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 6.0, | |
| "learning_rate": 0.00014, | |
| "loss": 0.0001, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 7.0, | |
| "learning_rate": 0.00013000000000000002, | |
| "loss": 0.0, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 8.0, | |
| "learning_rate": 0.00012, | |
| "loss": 0.0, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 9.0, | |
| "learning_rate": 0.00011000000000000002, | |
| "loss": 0.0, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "learning_rate": 0.0001, | |
| "loss": 0.0, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 10.0, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.9386414885520935, | |
| "eval_runtime": 0.2708, | |
| "eval_samples_per_second": 59.076, | |
| "eval_steps_per_second": 7.385, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 11.0, | |
| "learning_rate": 9e-05, | |
| "loss": 0.0, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 12.0, | |
| "learning_rate": 8e-05, | |
| "loss": 0.0, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 13.0, | |
| "learning_rate": 7e-05, | |
| "loss": 0.0, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 14.0, | |
| "learning_rate": 6e-05, | |
| "loss": 0.0, | |
| "step": 140 | |
| }, | |
| { | |
| "epoch": 15.0, | |
| "learning_rate": 5e-05, | |
| "loss": 0.0, | |
| "step": 150 | |
| }, | |
| { | |
| "epoch": 16.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0, | |
| "step": 160 | |
| }, | |
| { | |
| "epoch": 17.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0, | |
| "step": 170 | |
| }, | |
| { | |
| "epoch": 18.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0, | |
| "step": 180 | |
| }, | |
| { | |
| "epoch": 19.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0, | |
| "step": 190 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "eval_accuracy": 0.8125, | |
| "eval_loss": 0.9370558857917786, | |
| "eval_runtime": 0.2545, | |
| "eval_samples_per_second": 62.865, | |
| "eval_steps_per_second": 7.858, | |
| "step": 200 | |
| }, | |
| { | |
| "epoch": 20.0, | |
| "step": 200, | |
| "total_flos": 6.19947029495808e+16, | |
| "train_loss": 4.7035366296768185e-05, | |
| "train_runtime": 37.8678, | |
| "train_samples_per_second": 21.126, | |
| "train_steps_per_second": 5.282 | |
| } | |
| ], | |
| "max_steps": 200, | |
| "num_train_epochs": 20, | |
| "total_flos": 6.19947029495808e+16, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |