| { |
| "task": "forgetting_action_policy", |
| "epoch_stats": [ |
| { |
| "epoch": 1, |
| "train_loss": 1.2100253536409593, |
| "eval_macro_f1": 0.8943120476934178, |
| "eval_accuracy": 0.8914 |
| }, |
| { |
| "epoch": 2, |
| "train_loss": 0.6452919559225608, |
| "eval_macro_f1": 0.9476649277600788, |
| "eval_accuracy": 0.9484 |
| }, |
| { |
| "epoch": 3, |
| "train_loss": 0.482522361606962, |
| "eval_macro_f1": 0.9696191896424656, |
| "eval_accuracy": 0.9696 |
| }, |
| { |
| "epoch": 4, |
| "train_loss": 0.3768140496231193, |
| "eval_macro_f1": 0.9718038787431208, |
| "eval_accuracy": 0.9718 |
| }, |
| { |
| "epoch": 5, |
| "train_loss": 0.2870782903271907, |
| "eval_macro_f1": 0.9796027083744441, |
| "eval_accuracy": 0.9796 |
| }, |
| { |
| "epoch": 6, |
| "train_loss": 0.20888356409232609, |
| "eval_macro_f1": 0.9825928670083165, |
| "eval_accuracy": 0.9826 |
| }, |
| { |
| "epoch": 7, |
| "train_loss": 0.15703109322336797, |
| "eval_macro_f1": 0.9839922007228425, |
| "eval_accuracy": 0.984 |
| }, |
| { |
| "epoch": 8, |
| "train_loss": 0.11633934278838558, |
| "eval_macro_f1": 0.9835931767915431, |
| "eval_accuracy": 0.9836 |
| } |
| ], |
| "training_summary": { |
| "actual_epochs": 8, |
| "best_epoch": 7, |
| "early_stopped": false, |
| "backbone_model_name": "microsoft/deberta-v3-base", |
| "tokenizer_name": "microsoft/deberta-v3-base", |
| "selection_metric": "macro_f1", |
| "selection_value": 0.9839922007228425 |
| }, |
| "calibration": { |
| "method": "temperature_grid_search", |
| "rows": 5000, |
| "temperature": 2.0, |
| "loss": 0.08134587520392533 |
| }, |
| "hf_model_dir": "/nvme/CognitiveMemoryLayer/packages/models/trained_models/forgetting_action_policy_hf" |
| } |
