ner-hw2

This model is a fine-tuned version of BAAI/bge-small-en-v1.5 on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.1126
• Precision: 0.9188
• Recall: 0.9349
• F1: 0.9268
• Accuracy: 0.9836

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: 5e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED 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	Accuracy
0.0767	1.0	1252	0.0880	0.8771	0.9042	0.8905	0.9783
0.0488	2.0	2504	0.0908	0.8907	0.9152	0.9028	0.9790
0.0357	3.0	3756	0.0889	0.9048	0.9194	0.9120	0.9805
0.0263	4.0	5008	0.0828	0.8966	0.9300	0.9130	0.9815
0.0176	5.0	6260	0.0842	0.9059	0.9280	0.9168	0.9823
0.0146	6.0	7512	0.0958	0.9084	0.9194	0.9139	0.9824
0.0111	7.0	8764	0.0993	0.9119	0.9337	0.9227	0.9829
0.0086	8.0	10016	0.1005	0.9137	0.9315	0.9225	0.9822
0.0063	9.0	11268	0.1019	0.9073	0.9307	0.9188	0.9817
0.0046	10.0	12520	0.1027	0.9184	0.9297	0.9240	0.9835
0.0044	11.0	13772	0.1092	0.9134	0.9298	0.9215	0.9830
0.0022	12.0	15024	0.1124	0.9139	0.9286	0.9212	0.9830
0.0025	13.0	16276	0.1150	0.9199	0.9334	0.9266	0.9837
0.0019	14.0	17528	0.1125	0.9207	0.9355	0.9280	0.9838
0.0010	15.0	18780	0.1126	0.9188	0.9349	0.9268	0.9836

Framework versions

• Transformers 5.19.0
• Pytorch 2.11.0+cu128
• Datasets 5.1.0
• Tokenizers 0.23.2