model_2_stage2-seed_123

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: 2.7954
• Precision: 0.6352
• Recall: 0.7319
• F1: 0.6801

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: 123
• 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
0.5075	1.0	447	0.6591	0.6296	0.6159	0.6227
0.4241	2.0	894	0.8133	0.6012	0.7536	0.6688
0.3651	3.0	1341	0.8972	0.6092	0.7174	0.6589
0.3389	4.0	1788	1.3235	0.6040	0.7572	0.6720
0.2572	5.0	2235	1.2850	0.6378	0.7464	0.6878
0.187	6.0	2682	1.4055	0.6114	0.7754	0.6837
0.1456	7.0	3129	1.8037	0.6464	0.6558	0.6511
0.1386	8.0	3576	1.8962	0.6181	0.6920	0.6530
0.1003	9.0	4023	2.1076	0.6198	0.7029	0.6587
0.0738	10.0	4470	2.4260	0.6463	0.7283	0.6848
0.0233	11.0	4917	2.5047	0.6242	0.7464	0.6799
0.0677	12.0	5364	2.6329	0.6238	0.7029	0.6610
0.0249	13.0	5811	2.5839	0.6429	0.7174	0.6781
0.0249	14.0	6258	2.7944	0.6347	0.7428	0.6845
0.0228	15.0	6705	2.7954	0.6352	0.7319	0.6801

Framework versions

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