| { | |
| "best_metric": 0.87, | |
| "best_model_checkpoint": "fine-tune-vit-cifar100\\checkpoint-2816", | |
| "epoch": 5.0, | |
| "global_step": 3520, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.71, | |
| "learning_rate": 0.00019715909090909094, | |
| "loss": 2.6561, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.824, | |
| "eval_loss": 0.7328090071678162, | |
| "eval_runtime": 24.9237, | |
| "eval_samples_per_second": 200.612, | |
| "eval_steps_per_second": 3.17, | |
| "step": 704 | |
| }, | |
| { | |
| "epoch": 1.42, | |
| "learning_rate": 0.0001943181818181818, | |
| "loss": 1.2132, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8446, | |
| "eval_loss": 0.5557541251182556, | |
| "eval_runtime": 24.0685, | |
| "eval_samples_per_second": 207.74, | |
| "eval_steps_per_second": 3.282, | |
| "step": 1408 | |
| }, | |
| { | |
| "epoch": 2.13, | |
| "learning_rate": 0.00019147727272727274, | |
| "loss": 1.0151, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.84, | |
| "learning_rate": 0.00018863636363636364, | |
| "loss": 0.9014, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8508, | |
| "eval_loss": 0.5289514660835266, | |
| "eval_runtime": 23.869, | |
| "eval_samples_per_second": 209.477, | |
| "eval_steps_per_second": 3.31, | |
| "step": 2112 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 0.00018579545454545454, | |
| "loss": 0.8114, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.87, | |
| "eval_loss": 0.476001501083374, | |
| "eval_runtime": 24.1421, | |
| "eval_samples_per_second": 207.107, | |
| "eval_steps_per_second": 3.272, | |
| "step": 2816 | |
| }, | |
| { | |
| "epoch": 4.26, | |
| "learning_rate": 0.00018295454545454547, | |
| "loss": 0.7639, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.97, | |
| "learning_rate": 0.00018011363636363638, | |
| "loss": 0.7272, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8648, | |
| "eval_loss": 0.46676263213157654, | |
| "eval_runtime": 24.1887, | |
| "eval_samples_per_second": 206.708, | |
| "eval_steps_per_second": 3.266, | |
| "step": 3520 | |
| } | |
| ], | |
| "max_steps": 35200, | |
| "num_train_epochs": 50, | |
| "total_flos": 1.74510121863168e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |