| { | |
| "best_metric": 0.0645078495144844, | |
| "best_model_checkpoint": "./vit-base-beans-demo-v3/checkpoint-100", | |
| "epoch": 2.0, | |
| "global_step": 130, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.15, | |
| "learning_rate": 0.00018461538461538463, | |
| "loss": 0.8806, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.31, | |
| "learning_rate": 0.00016923076923076923, | |
| "loss": 0.4438, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 0.46, | |
| "learning_rate": 0.00015384615384615385, | |
| "loss": 0.3464, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 0.62, | |
| "learning_rate": 0.00013846153846153847, | |
| "loss": 0.3524, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 0.77, | |
| "learning_rate": 0.0001230769230769231, | |
| "loss": 0.2376, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 0.92, | |
| "learning_rate": 0.0001076923076923077, | |
| "loss": 0.1484, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 1.08, | |
| "learning_rate": 9.230769230769232e-05, | |
| "loss": 0.0672, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 1.23, | |
| "learning_rate": 7.692307692307693e-05, | |
| "loss": 0.1207, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 1.38, | |
| "learning_rate": 6.153846153846155e-05, | |
| "loss": 0.0542, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "learning_rate": 4.615384615384616e-05, | |
| "loss": 0.0397, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.54, | |
| "eval_accuracy": 0.9849624060150376, | |
| "eval_loss": 0.0645078495144844, | |
| "eval_runtime": 4.5427, | |
| "eval_samples_per_second": 29.278, | |
| "eval_steps_per_second": 3.742, | |
| "step": 100 | |
| }, | |
| { | |
| "epoch": 1.69, | |
| "learning_rate": 3.0769230769230774e-05, | |
| "loss": 0.0978, | |
| "step": 110 | |
| }, | |
| { | |
| "epoch": 1.85, | |
| "learning_rate": 1.5384615384615387e-05, | |
| "loss": 0.0357, | |
| "step": 120 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0721, | |
| "step": 130 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "step": 130, | |
| "total_flos": 0.0, | |
| "train_loss": 0.22280787963133591, | |
| "train_runtime": 105.6076, | |
| "train_samples_per_second": 19.582, | |
| "train_steps_per_second": 1.231 | |
| } | |
| ], | |
| "max_steps": 130, | |
| "num_train_epochs": 2, | |
| "total_flos": 0.0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |