README.md

metadata

license: apache-2.0
base_model: bert-base-cased
tags:
  - generated_from_trainer
metrics:
  - precision
  - recall
  - f1
  - accuracy
model-index:
  - name: bert-finetuned-ner
    results: []

bert-finetuned-ner

This model is a fine-tuned version of bert-base-cased on the None dataset. It achieves the following results on the evaluation set:

• Loss: 0.5707
• Precision: 0.8258
• Recall: 0.8753
• F1: 0.8498
• Accuracy: 0.9254

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: 1e-05
• train_batch_size: 8
• eval_batch_size: 8
• seed: 42
• optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08
• lr_scheduler_type: linear
• num_epochs: 20

Training results

Training Loss	Epoch	Step	Validation Loss	Precision	Recall	F1	Accuracy
0.1966	1.0	680	0.3961	0.7705	0.8096	0.7896	0.8971
0.1487	2.0	1360	0.4435	0.7607	0.8258	0.7919	0.8950
0.1258	3.0	2040	0.4180	0.7727	0.8450	0.8072	0.9052
0.1163	4.0	2720	0.4227	0.7890	0.8418	0.8145	0.9078
0.1019	5.0	3400	0.4721	0.7664	0.8630	0.8118	0.9068
0.0763	6.0	4080	0.4444	0.8033	0.8542	0.8280	0.9172
0.0641	7.0	4760	0.4551	0.8189	0.8550	0.8366	0.9179
0.0615	8.0	5440	0.4615	0.8167	0.8640	0.8397	0.9202
0.0468	9.0	6120	0.4922	0.8104	0.8668	0.8376	0.9205
0.0381	10.0	6800	0.4991	0.8213	0.8642	0.8422	0.9210
0.0381	11.0	7480	0.4950	0.8298	0.8593	0.8443	0.9230
0.0289	12.0	8160	0.5273	0.8220	0.8717	0.8461	0.9232
0.026	13.0	8840	0.5366	0.8207	0.8700	0.8446	0.9230
0.0227	14.0	9520	0.5466	0.8240	0.8698	0.8463	0.9230
0.02	15.0	10200	0.5530	0.8250	0.8710	0.8474	0.9247
0.0173	16.0	10880	0.5571	0.8235	0.8726	0.8473	0.9244
0.0152	17.0	11560	0.5551	0.8282	0.8700	0.8486	0.9248
0.0146	18.0	12240	0.5686	0.8270	0.8727	0.8492	0.9246
0.0142	19.0	12920	0.5687	0.8258	0.8753	0.8498	0.9259
0.0135	20.0	13600	0.5707	0.8258	0.8753	0.8498	0.9254

Framework versions

• Transformers 4.42.4
• Pytorch 2.3.1+cu121
• Datasets 2.20.0
• Tokenizers 0.19.1