model_2_stage2-seed_42

This model is a fine-tuned version of maud-dr/model_2_stage1 on the None dataset. It achieves the following results on the evaluation set:

• Loss: 2.9477
• Precision: 0.6407
• Recall: 0.6848
• F1: 0.6620

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

• learning_rate: 0.0003
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
• lr_scheduler_type: linear
• num_epochs: 15

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1
0.6494	1.0	447	0.7733	0.5673	0.7790	0.6565
0.5791	2.0	894	0.6312	0.6076	0.8080	0.6936
0.5043	3.0	1341	0.7073	0.6171	0.7065	0.6588
0.4117	4.0	1788	0.7908	0.6198	0.75	0.6787
0.3571	5.0	2235	1.0448	0.6017	0.75	0.6677
0.3421	6.0	2682	1.1355	0.6176	0.7138	0.6622
0.2765	7.0	3129	1.2417	0.6512	0.6630	0.6571
0.2304	8.0	3576	1.5647	0.6329	0.7246	0.6757
0.1435	9.0	4023	2.0734	0.6378	0.7210	0.6769
0.1275	10.0	4470	2.0432	0.6403	0.7029	0.6701
0.1361	11.0	4917	2.2470	0.6403	0.7029	0.6701
0.0658	12.0	5364	2.6546	0.6314	0.7138	0.6701
0.0626	13.0	5811	2.7080	0.6431	0.6920	0.6667
0.0396	14.0	6258	2.9436	0.6201	0.6920	0.6541
0.0475	15.0	6705	2.9477	0.6407	0.6848	0.6620

Framework versions

• Transformers 4.52.0.dev0
• Pytorch 2.7.0+cu126
• Datasets 3.6.0
• Tokenizers 0.21.1