distilbert-finetuned-ner_ep

This model is a fine-tuned version of leo1234messi/distilbert-finetuned-ner on an unknown dataset. It achieves the following results on the evaluation set:

• Loss: 0.2936
• Precision: 0.8886
• Recall: 0.8859
• F1: 0.8873
• Accuracy: 0.9418

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: 2e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Use 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: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
No log	1.0	42	0.2409	0.8208	0.8368	0.8287	0.9136
No log	2.0	84	0.1895	0.8633	0.8689	0.8661	0.9356
No log	3.0	126	0.1922	0.8591	0.8739	0.8664	0.9344
No log	4.0	168	0.1992	0.8887	0.8854	0.8870	0.9421
No log	5.0	210	0.2143	0.8670	0.8779	0.8724	0.9374
No log	6.0	252	0.2158	0.8732	0.8771	0.8751	0.9388
No log	7.0	294	0.2585	0.9011	0.8902	0.8956	0.9444
No log	8.0	336	0.2341	0.8903	0.8857	0.8880	0.9436
No log	9.0	378	0.2615	0.8935	0.8859	0.8897	0.9431
No log	10.0	420	0.2524	0.8872	0.8839	0.8855	0.9434
No log	11.0	462	0.2736	0.8845	0.8826	0.8835	0.9409
0.0767	12.0	504	0.2821	0.8857	0.8824	0.8840	0.9406
0.0767	13.0	546	0.2726	0.8868	0.8880	0.8874	0.9418
0.0767	14.0	588	0.2796	0.8889	0.8859	0.8874	0.9416
0.0767	15.0	630	0.2857	0.8905	0.8846	0.8875	0.9413
0.0767	16.0	672	0.2884	0.8873	0.8833	0.8853	0.9412
0.0767	17.0	714	0.2917	0.8880	0.8839	0.8859	0.9408
0.0767	18.0	756	0.2955	0.8908	0.8859	0.8883	0.9417
0.0767	19.0	798	0.2950	0.8892	0.8852	0.8872	0.9416
0.0767	20.0	840	0.2936	0.8886	0.8859	0.8873	0.9418

Framework versions

• Transformers 4.47.0
• Pytorch 2.5.1+cu121
• Tokenizers 0.21.0