| { | |
| "best_metric": 0.9531666666666667, | |
| "best_model_checkpoint": "fine-tune-vit-L-fashionmnist\\checkpoint-4220", | |
| "epoch": 5.0, | |
| "global_step": 4220, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.59, | |
| "learning_rate": 0.00017630331753554503, | |
| "loss": 0.6404, | |
| "step": 500 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9026666666666666, | |
| "eval_loss": 0.26137667894363403, | |
| "eval_runtime": 70.6906, | |
| "eval_samples_per_second": 84.877, | |
| "eval_steps_per_second": 1.33, | |
| "step": 844 | |
| }, | |
| { | |
| "epoch": 1.18, | |
| "learning_rate": 0.00015260663507109004, | |
| "loss": 0.495, | |
| "step": 1000 | |
| }, | |
| { | |
| "epoch": 1.78, | |
| "learning_rate": 0.0001289099526066351, | |
| "loss": 0.4369, | |
| "step": 1500 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9141666666666667, | |
| "eval_loss": 0.23407751321792603, | |
| "eval_runtime": 73.6684, | |
| "eval_samples_per_second": 81.446, | |
| "eval_steps_per_second": 1.276, | |
| "step": 1688 | |
| }, | |
| { | |
| "epoch": 2.37, | |
| "learning_rate": 0.0001052132701421801, | |
| "loss": 0.3966, | |
| "step": 2000 | |
| }, | |
| { | |
| "epoch": 2.96, | |
| "learning_rate": 8.151658767772512e-05, | |
| "loss": 0.3582, | |
| "step": 2500 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9393333333333334, | |
| "eval_loss": 0.15943674743175507, | |
| "eval_runtime": 67.7669, | |
| "eval_samples_per_second": 88.539, | |
| "eval_steps_per_second": 1.387, | |
| "step": 2532 | |
| }, | |
| { | |
| "epoch": 3.55, | |
| "learning_rate": 5.781990521327014e-05, | |
| "loss": 0.3155, | |
| "step": 3000 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9461666666666667, | |
| "eval_loss": 0.14314088225364685, | |
| "eval_runtime": 69.4201, | |
| "eval_samples_per_second": 86.43, | |
| "eval_steps_per_second": 1.354, | |
| "step": 3376 | |
| }, | |
| { | |
| "epoch": 4.15, | |
| "learning_rate": 3.412322274881517e-05, | |
| "loss": 0.2971, | |
| "step": 3500 | |
| }, | |
| { | |
| "epoch": 4.74, | |
| "learning_rate": 1.0426540284360189e-05, | |
| "loss": 0.2572, | |
| "step": 4000 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9531666666666667, | |
| "eval_loss": 0.12807205319404602, | |
| "eval_runtime": 68.1462, | |
| "eval_samples_per_second": 88.046, | |
| "eval_steps_per_second": 1.379, | |
| "step": 4220 | |
| } | |
| ], | |
| "max_steps": 4220, | |
| "num_train_epochs": 5, | |
| "total_flos": 7.396422817738752e+19, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |