|
--- |
|
license: mit |
|
base_model: roberta-base |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- f1 |
|
- accuracy |
|
model-index: |
|
- name: ner-fine-tune-roberta-new |
|
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-roberta-new |
|
|
|
This model is a fine-tuned version of [roberta-base](https://huggingface.co/roberta-base) on the None dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3320 |
|
- Precision: 0.2696 |
|
- Recall: 0.3767 |
|
- F1: 0.3143 |
|
- Accuracy: 0.9389 |
|
|
|
## 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 | 0.2264 | 0.0 | 0.0 | 0.0 | 0.9542 | |
|
| No log | 2.0 | 244 | 0.1874 | 0.2348 | 0.1349 | 0.1713 | 0.9571 | |
|
| No log | 3.0 | 366 | 0.1739 | 0.2420 | 0.2279 | 0.2347 | 0.9528 | |
|
| No log | 4.0 | 488 | 0.1656 | 0.1939 | 0.2233 | 0.2076 | 0.9472 | |
|
| 0.2548 | 5.0 | 610 | 0.1740 | 0.3243 | 0.2767 | 0.2986 | 0.9573 | |
|
| 0.2548 | 6.0 | 732 | 0.2087 | 0.2562 | 0.2628 | 0.2595 | 0.9464 | |
|
| 0.2548 | 7.0 | 854 | 0.1921 | 0.2773 | 0.2953 | 0.2860 | 0.9495 | |
|
| 0.2548 | 8.0 | 976 | 0.2038 | 0.2602 | 0.3860 | 0.3109 | 0.9397 | |
|
| 0.0748 | 9.0 | 1098 | 0.2324 | 0.2371 | 0.3093 | 0.2684 | 0.9398 | |
|
| 0.0748 | 10.0 | 1220 | 0.2329 | 0.2852 | 0.3442 | 0.3119 | 0.9436 | |
|
| 0.0748 | 11.0 | 1342 | 0.2670 | 0.2521 | 0.3535 | 0.2943 | 0.9356 | |
|
| 0.0748 | 12.0 | 1464 | 0.2607 | 0.2509 | 0.3186 | 0.2807 | 0.9395 | |
|
| 0.033 | 13.0 | 1586 | 0.2645 | 0.2655 | 0.3791 | 0.3123 | 0.9359 | |
|
| 0.033 | 14.0 | 1708 | 0.2947 | 0.2838 | 0.4442 | 0.3463 | 0.9398 | |
|
| 0.033 | 15.0 | 1830 | 0.2807 | 0.2945 | 0.3349 | 0.3134 | 0.9451 | |
|
| 0.033 | 16.0 | 1952 | 0.2990 | 0.2910 | 0.3302 | 0.3094 | 0.9448 | |
|
| 0.0181 | 17.0 | 2074 | 0.2915 | 0.2799 | 0.3651 | 0.3169 | 0.9425 | |
|
| 0.0181 | 18.0 | 2196 | 0.2853 | 0.2868 | 0.3535 | 0.3167 | 0.9424 | |
|
| 0.0181 | 19.0 | 2318 | 0.2991 | 0.2918 | 0.3814 | 0.3306 | 0.9440 | |
|
| 0.0181 | 20.0 | 2440 | 0.2863 | 0.2762 | 0.3744 | 0.3179 | 0.9408 | |
|
| 0.0111 | 21.0 | 2562 | 0.3280 | 0.2796 | 0.3628 | 0.3158 | 0.9409 | |
|
| 0.0111 | 22.0 | 2684 | 0.3135 | 0.2772 | 0.3372 | 0.3043 | 0.9431 | |
|
| 0.0111 | 23.0 | 2806 | 0.3282 | 0.2632 | 0.3698 | 0.3075 | 0.9404 | |
|
| 0.0111 | 24.0 | 2928 | 0.3306 | 0.2597 | 0.3884 | 0.3113 | 0.9369 | |
|
| 0.0078 | 25.0 | 3050 | 0.3135 | 0.2743 | 0.3605 | 0.3116 | 0.9414 | |
|
| 0.0078 | 26.0 | 3172 | 0.3320 | 0.2646 | 0.3791 | 0.3117 | 0.9374 | |
|
| 0.0078 | 27.0 | 3294 | 0.3273 | 0.2659 | 0.3791 | 0.3126 | 0.9381 | |
|
| 0.0078 | 28.0 | 3416 | 0.3290 | 0.2616 | 0.3674 | 0.3056 | 0.9380 | |
|
| 0.0055 | 29.0 | 3538 | 0.3366 | 0.2656 | 0.3860 | 0.3147 | 0.9372 | |
|
| 0.0055 | 30.0 | 3660 | 0.3320 | 0.2696 | 0.3767 | 0.3143 | 0.9389 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.34.1 |
|
- Pytorch 2.1.0+cu118 |
|
- Datasets 2.14.6 |
|
- Tokenizers 0.14.1 |
|
|
