|
--- |
|
license: mit |
|
base_model: dslim/bert-base-NER |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: ner-fine-tune-bert-ner |
|
results: [] |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# ner-fine-tune-bert-ner |
|
|
|
This model is a fine-tuned version of [dslim/bert-base-NER](https://huggingface.co/dslim/bert-base-NER) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3662 |
|
- Precision: 0.2383 |
|
- Recall: 0.2818 |
|
- F1: 0.2582 |
|
- Accuracy: 0.9406 |
|
|
|
## 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: 30 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Accuracy | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:------:|:--------:| |
|
| No log | 1.0 | 122 | 0.2295 | 0.1255 | 0.0716 | 0.0912 | 0.9514 | |
|
| No log | 2.0 | 244 | 0.2152 | 0.2022 | 0.1270 | 0.1560 | 0.9514 | |
|
| No log | 3.0 | 366 | 0.2044 | 0.1696 | 0.1547 | 0.1618 | 0.9497 | |
|
| No log | 4.0 | 488 | 0.2269 | 0.1980 | 0.1363 | 0.1614 | 0.9536 | |
|
| 0.2142 | 5.0 | 610 | 0.2335 | 0.1931 | 0.1547 | 0.1718 | 0.9521 | |
|
| 0.2142 | 6.0 | 732 | 0.2516 | 0.1959 | 0.1778 | 0.1864 | 0.9491 | |
|
| 0.2142 | 7.0 | 854 | 0.2446 | 0.2565 | 0.2517 | 0.2541 | 0.9542 | |
|
| 0.2142 | 8.0 | 976 | 0.2527 | 0.2273 | 0.2656 | 0.2449 | 0.9481 | |
|
| 0.0658 | 9.0 | 1098 | 0.2724 | 0.2459 | 0.2055 | 0.2239 | 0.9526 | |
|
| 0.0658 | 10.0 | 1220 | 0.2620 | 0.2895 | 0.2748 | 0.2820 | 0.9549 | |
|
| 0.0658 | 11.0 | 1342 | 0.2846 | 0.2102 | 0.2748 | 0.2382 | 0.9416 | |
|
| 0.0658 | 12.0 | 1464 | 0.2943 | 0.2292 | 0.2610 | 0.2441 | 0.9450 | |
|
| 0.0273 | 13.0 | 1586 | 0.3154 | 0.2064 | 0.2679 | 0.2332 | 0.9381 | |
|
| 0.0273 | 14.0 | 1708 | 0.3097 | 0.2254 | 0.2217 | 0.2235 | 0.9464 | |
|
| 0.0273 | 15.0 | 1830 | 0.3313 | 0.2375 | 0.2517 | 0.2444 | 0.9426 | |
|
| 0.0273 | 16.0 | 1952 | 0.3256 | 0.2098 | 0.2864 | 0.2422 | 0.9361 | |
|
| 0.0155 | 17.0 | 2074 | 0.3333 | 0.2162 | 0.2656 | 0.2383 | 0.9393 | |
|
| 0.0155 | 18.0 | 2196 | 0.3073 | 0.2446 | 0.2864 | 0.2638 | 0.9449 | |
|
| 0.0155 | 19.0 | 2318 | 0.3241 | 0.2418 | 0.2725 | 0.2562 | 0.9437 | |
|
| 0.0155 | 20.0 | 2440 | 0.3348 | 0.2338 | 0.2587 | 0.2456 | 0.9446 | |
|
| 0.0091 | 21.0 | 2562 | 0.3595 | 0.234 | 0.2702 | 0.2508 | 0.9402 | |
|
| 0.0091 | 22.0 | 2684 | 0.3658 | 0.2263 | 0.2818 | 0.2510 | 0.9387 | |
|
| 0.0091 | 23.0 | 2806 | 0.3495 | 0.2391 | 0.2794 | 0.2577 | 0.9419 | |
|
| 0.0091 | 24.0 | 2928 | 0.3545 | 0.2398 | 0.2841 | 0.2600 | 0.9409 | |
|
| 0.0066 | 25.0 | 3050 | 0.3557 | 0.2309 | 0.2864 | 0.2557 | 0.9402 | |
|
| 0.0066 | 26.0 | 3172 | 0.3498 | 0.2449 | 0.2748 | 0.2590 | 0.9432 | |
|
| 0.0066 | 27.0 | 3294 | 0.3586 | 0.2375 | 0.2841 | 0.2587 | 0.9416 | |
|
| 0.0066 | 28.0 | 3416 | 0.3676 | 0.2389 | 0.2725 | 0.2546 | 0.9417 | |
|
| 0.005 | 29.0 | 3538 | 0.3663 | 0.2412 | 0.2864 | 0.2619 | 0.9404 | |
|
| 0.005 | 30.0 | 3660 | 0.3662 | 0.2383 | 0.2818 | 0.2582 | 0.9406 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.34.1 |
|
- Pytorch 2.1.0+cu118 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
