| { | |
| "best_metric": 0.870096790626592, | |
| "best_model_checkpoint": "./outputs/xlnet-base-cased/fine_tuned_models/checkpoint-49088", | |
| "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": 0.5395, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.8585838003056546, | |
| "eval_loss": 0.3862307071685791, | |
| "eval_runtime": 36.1864, | |
| "eval_samples_per_second": 271.234, | |
| "eval_steps_per_second": 33.908, | |
| "step": 12272 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 1.3041445270988312e-05, | |
| "loss": 0.3458, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.8625573102394295, | |
| "eval_loss": 0.3799586296081543, | |
| "eval_runtime": 36.1789, | |
| "eval_samples_per_second": 271.291, | |
| "eval_steps_per_second": 33.915, | |
| "step": 24544 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 8.694296847325541e-06, | |
| "loss": 0.2464, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.867142129393785, | |
| "eval_loss": 0.4079091250896454, | |
| "eval_runtime": 36.209, | |
| "eval_samples_per_second": 271.065, | |
| "eval_steps_per_second": 33.887, | |
| "step": 36816 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 4.3471484236627705e-06, | |
| "loss": 0.1763, | |
| "step": 49088 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.870096790626592, | |
| "eval_loss": 0.4703195095062256, | |
| "eval_runtime": 36.1832, | |
| "eval_samples_per_second": 271.258, | |
| "eval_steps_per_second": 33.911, | |
| "step": 49088 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.1293, | |
| "step": 61360 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.8691798267957208, | |
| "eval_loss": 0.5571475028991699, | |
| "eval_runtime": 36.1737, | |
| "eval_samples_per_second": 271.329, | |
| "eval_steps_per_second": 33.92, | |
| "step": 61360 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 61360, | |
| "total_flos": 1.769024893707648e+17, | |
| "train_loss": 0.28748211518726574, | |
| "train_runtime": 20307.0185, | |
| "train_samples_per_second": 96.691, | |
| "train_steps_per_second": 3.022 | |
| } | |
| ], | |
| "max_steps": 61360, | |
| "num_train_epochs": 5, | |
| "total_flos": 1.769024893707648e+17, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |