| { | |
| "best_metric": 0.9003506721215664, | |
| "best_model_checkpoint": "output/roberta-base-bne-capitel-iris-ner/iris_16_0.00005_07-17-23_22-27/checkpoint-3540", | |
| "epoch": 5.0, | |
| "global_step": 3540, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 1.0, | |
| "learning_rate": 4e-05, | |
| "loss": 0.0969, | |
| "step": 708 | |
| }, | |
| { | |
| "epoch": 1.0, | |
| "eval_accuracy": 0.9882810956088692, | |
| "eval_f1": 0.8684808352264488, | |
| "eval_loss": 0.04088732972741127, | |
| "eval_precision": 0.8588910133843213, | |
| "eval_recall": 0.8782872226024049, | |
| "eval_runtime": 34.6042, | |
| "eval_samples_per_second": 218.182, | |
| "step": 708 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "learning_rate": 3e-05, | |
| "loss": 0.0253, | |
| "step": 1416 | |
| }, | |
| { | |
| "epoch": 2.0, | |
| "eval_accuracy": 0.9890666376823051, | |
| "eval_f1": 0.8800116352353711, | |
| "eval_loss": 0.03929711878299713, | |
| "eval_precision": 0.8728601654164262, | |
| "eval_recall": 0.8872812591651188, | |
| "eval_runtime": 34.2725, | |
| "eval_samples_per_second": 220.293, | |
| "step": 1416 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "learning_rate": 2e-05, | |
| "loss": 0.0117, | |
| "step": 2124 | |
| }, | |
| { | |
| "epoch": 3.0, | |
| "eval_accuracy": 0.9898768823366665, | |
| "eval_f1": 0.892319680514294, | |
| "eval_loss": 0.03888658434152603, | |
| "eval_precision": 0.8890722049689441, | |
| "eval_recall": 0.8955909668589305, | |
| "eval_runtime": 34.6426, | |
| "eval_samples_per_second": 217.94, | |
| "step": 2124 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "learning_rate": 1e-05, | |
| "loss": 0.0054, | |
| "step": 2832 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "eval_accuracy": 0.9904549227303269, | |
| "eval_f1": 0.8978559968885217, | |
| "eval_loss": 0.040431126952171326, | |
| "eval_precision": 0.893036750483559, | |
| "eval_recall": 0.9027275393489099, | |
| "eval_runtime": 34.4446, | |
| "eval_samples_per_second": 219.192, | |
| "step": 2832 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "learning_rate": 0.0, | |
| "loss": 0.0026, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "eval_accuracy": 0.9905438520216592, | |
| "eval_f1": 0.9003506721215664, | |
| "eval_loss": 0.04321206733584404, | |
| "eval_precision": 0.8971173444627778, | |
| "eval_recall": 0.9036073907517841, | |
| "eval_runtime": 34.2514, | |
| "eval_samples_per_second": 220.429, | |
| "step": 3540 | |
| }, | |
| { | |
| "epoch": 5.0, | |
| "step": 3540, | |
| "total_flos": 0, | |
| "train_runtime": 1076.7664, | |
| "train_samples_per_second": 3.288 | |
| } | |
| ], | |
| "max_steps": 3540, | |
| "num_train_epochs": 5, | |
| "total_flos": 0, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |