| { | |
| "best_metric": 0.8705043301069791, | |
| "best_model_checkpoint": "./outputs/xlnet-base-cased/fine_tuned_models/checkpoint-61360", | |
| "epoch": 5.0, | |
| "global_step": 61360, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 1.7388593694651082e-05, | |
| "loss": 0.5438, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8439123790117168, | |
| "eval_loss": 0.4061104655265808, | |
| "eval_runtime": 36.1298, | |
| "eval_samples_per_second": 271.66, | |
| "eval_steps_per_second": 33.961, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.3041445270988312e-05, | |
| "loss": 0.3472, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.869077941925624, | |
| "eval_loss": 0.37025684118270874, | |
| "eval_runtime": 36.0225, | |
| "eval_samples_per_second": 272.468, | |
| "eval_steps_per_second": 34.062, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.694296847325541e-06, | |
| "loss": 0.2487, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.8699949057564952, | |
| "eval_loss": 0.38715142011642456, | |
| "eval_runtime": 36.0648, | |
| "eval_samples_per_second": 272.149, | |
| "eval_steps_per_second": 34.022, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.3471484236627705e-06, | |
| "loss": 0.1762, | |
| "step": 49088 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.8680590932246561, | |
| "eval_loss": 0.46673768758773804, | |
| "eval_runtime": 36.0869, | |
| "eval_samples_per_second": 271.983, | |
| "eval_steps_per_second": 34.001, | |
| "step": 49088 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.1306, | |
| "step": 61360 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8705043301069791, | |
| "eval_loss": 0.5481665134429932, | |
| "eval_runtime": 36.1255, | |
| "eval_samples_per_second": 271.692, | |
| "eval_steps_per_second": 33.965, | |
| "step": 61360 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 61360, | |
| "total_flos": 1.769024893707648e+17, | |
| "train_loss": 0.2893029940330376, | |
| "train_runtime": 20201.1589, | |
| "train_samples_per_second": 97.198, | |
| "train_steps_per_second": 3.037 | |
| } | |
| ], | |
| "max_steps": 61360, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.769024893707648e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |