| { |
| "best_metric": 0.0044614835642278194, |
| "best_model_checkpoint": "./vit-base-beans/checkpoint-80", |
| "epoch": 4.0, |
| "global_step": 100, |
| "is_hyper_param_search": false, |
| "is_local_process_zero": true, |
| "is_world_process_zero": true, |
| "log_history": [ |
| { |
| "epoch": 0.4, |
| "learning_rate": 0.00018, |
| "loss": 0.244, |
| "step": 10 |
| }, |
| { |
| "epoch": 0.8, |
| "learning_rate": 0.00016, |
| "loss": 0.019, |
| "step": 20 |
| }, |
| { |
| "epoch": 1.2, |
| "learning_rate": 0.00014, |
| "loss": 0.0092, |
| "step": 30 |
| }, |
| { |
| "epoch": 1.6, |
| "learning_rate": 0.00012, |
| "loss": 0.0068, |
| "step": 40 |
| }, |
| { |
| "epoch": 1.6, |
| "eval_accuracy": 1.0, |
| "eval_loss": 0.006160833407193422, |
| "eval_runtime": 1.4649, |
| "eval_samples_per_second": 75.773, |
| "eval_steps_per_second": 9.557, |
| "step": 40 |
| }, |
| { |
| "epoch": 2.0, |
| "learning_rate": 0.0001, |
| "loss": 0.0058, |
| "step": 50 |
| }, |
| { |
| "epoch": 2.4, |
| "learning_rate": 8e-05, |
| "loss": 0.0051, |
| "step": 60 |
| }, |
| { |
| "epoch": 2.8, |
| "learning_rate": 6e-05, |
| "loss": 0.0048, |
| "step": 70 |
| }, |
| { |
| "epoch": 3.2, |
| "learning_rate": 4e-05, |
| "loss": 0.0045, |
| "step": 80 |
| }, |
| { |
| "epoch": 3.2, |
| "eval_accuracy": 1.0, |
| "eval_loss": 0.0044614835642278194, |
| "eval_runtime": 1.5025, |
| "eval_samples_per_second": 73.876, |
| "eval_steps_per_second": 9.318, |
| "step": 80 |
| }, |
| { |
| "epoch": 3.6, |
| "learning_rate": 2e-05, |
| "loss": 0.0044, |
| "step": 90 |
| }, |
| { |
| "epoch": 4.0, |
| "learning_rate": 0.0, |
| "loss": 0.0043, |
| "step": 100 |
| }, |
| { |
| "epoch": 4.0, |
| "step": 100, |
| "total_flos": 1.2057753584030515e+17, |
| "train_loss": 0.030779192969202996, |
| "train_runtime": 74.2102, |
| "train_samples_per_second": 20.967, |
| "train_steps_per_second": 1.348 |
| } |
| ], |
| "max_steps": 100, |
| "num_train_epochs": 4, |
| "total_flos": 1.2057753584030515e+17, |
| "trial_name": null, |
| "trial_params": null |
| } |
|
|
