<!-- This model card has been generated automatically according to the information the Trainer had access to. You should probably proofread and complete it, then remove this comment. -->
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: 0.0453
- Precision: 0.9275
- Recall: 0.9492
- F1: 0.9382
- Accuracy: 0.9934
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: Adam with betas=(0.9,0.999) and epsilon=1e-08
- lr_scheduler_type: linear
- num_epochs: 30
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy |
|---|---|---|---|---|---|---|---|
| No log | 1.0 | 407 | 0.0539 | 0.8283 | 0.8758 | 0.8514 | 0.9866 |
| 0.1524 | 2.0 | 814 | 0.0333 | 0.8931 | 0.9123 | 0.9026 | 0.9915 |
| 0.0381 | 3.0 | 1221 | 0.0345 | 0.8835 | 0.9280 | 0.9052 | 0.9906 |
| 0.0179 | 4.0 | 1628 | 0.0351 | 0.8890 | 0.9361 | 0.9119 | 0.9909 |
| 0.0089 | 5.0 | 2035 | 0.0310 | 0.9102 | 0.9372 | 0.9235 | 0.9924 |
| 0.0089 | 6.0 | 2442 | 0.0344 | 0.9198 | 0.9383 | 0.9289 | 0.9922 |
| 0.0057 | 7.0 | 2849 | 0.0331 | 0.9144 | 0.9448 | 0.9294 | 0.9931 |
| 0.0039 | 8.0 | 3256 | 0.0340 | 0.9144 | 0.9481 | 0.9309 | 0.9928 |
| 0.0027 | 9.0 | 3663 | 0.0423 | 0.9032 | 0.9481 | 0.9251 | 0.9921 |
| 0.0018 | 10.0 | 4070 | 0.0373 | 0.9047 | 0.9507 | 0.9271 | 0.9923 |
| 0.0018 | 11.0 | 4477 | 0.0448 | 0.8932 | 0.9474 | 0.9195 | 0.9910 |
| 0.0014 | 12.0 | 4884 | 0.0380 | 0.9079 | 0.9474 | 0.9272 | 0.9928 |
| 0.0015 | 13.0 | 5291 | 0.0360 | 0.9231 | 0.9474 | 0.9351 | 0.9936 |
| 0.0013 | 14.0 | 5698 | 0.0378 | 0.9243 | 0.9456 | 0.9348 | 0.9935 |
| 0.0013 | 15.0 | 6105 | 0.0414 | 0.9197 | 0.9496 | 0.9344 | 0.9930 |
| 0.0009 | 16.0 | 6512 | 0.0405 | 0.9202 | 0.9478 | 0.9338 | 0.9929 |
| 0.0009 | 17.0 | 6919 | 0.0385 | 0.9305 | 0.9441 | 0.9373 | 0.9933 |
| 0.0006 | 18.0 | 7326 | 0.0407 | 0.9285 | 0.9437 | 0.9360 | 0.9934 |
| 0.0009 | 19.0 | 7733 | 0.0428 | 0.9203 | 0.9488 | 0.9343 | 0.9929 |
| 0.0006 | 20.0 | 8140 | 0.0455 | 0.9232 | 0.9536 | 0.9382 | 0.9928 |
| 0.0004 | 21.0 | 8547 | 0.0462 | 0.9261 | 0.9529 | 0.9393 | 0.9930 |
| 0.0004 | 22.0 | 8954 | 0.0423 | 0.9359 | 0.9492 | 0.9425 | 0.9940 |
| 0.0005 | 23.0 | 9361 | 0.0446 | 0.9180 | 0.9529 | 0.9351 | 0.9931 |
| 0.0005 | 24.0 | 9768 | 0.0430 | 0.9361 | 0.9467 | 0.9413 | 0.9938 |
| 0.0002 | 25.0 | 10175 | 0.0436 | 0.9322 | 0.9496 | 0.9408 | 0.9935 |
| 0.0002 | 26.0 | 10582 | 0.0440 | 0.9275 | 0.9492 | 0.9382 | 0.9935 |
| 0.0002 | 27.0 | 10989 | 0.0450 | 0.9272 | 0.9488 | 0.9379 | 0.9932 |
| 0.0002 | 28.0 | 11396 | 0.0445 | 0.9304 | 0.9470 | 0.9386 | 0.9935 |
| 0.0003 | 29.0 | 11803 | 0.0449 | 0.9278 | 0.9481 | 0.9378 | 0.9934 |
| 0.0001 | 30.0 | 12210 | 0.0453 | 0.9275 | 0.9492 | 0.9382 | 0.9934 |
Framework versions
- Transformers 4.23.1
- Pytorch 1.12.1+cu113
- Datasets 2.6.1
- Tokenizers 0.13.1
