|
--- |
|
license: apache-2.0 |
|
tags: |
|
- generated_from_trainer |
|
datasets: |
|
- wnut_17 |
|
model-index: |
|
- name: fine_tune_bert_output |
|
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. --> |
|
|
|
# Bertweet-base finetuned on wnut17_ner |
|
|
|
This model is a fine-tuned version of [vinai/bertweet-base](https://huggingface.co/vinai/bertweet-base) on the [wnut_17](https://huggingface.co/datasets/wnut_17) dataset. |
|
|
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3239 |
|
- Overall Precision: 0.6913 |
|
- Overall Recall: 0.5914 |
|
- Overall F1: 0.6374 |
|
- Overall Accuracy: 0.9499 |
|
- Corporation F1: 0.2703 |
|
- Creative-work F1: 0.3636 |
|
- Group F1: 0.4030 |
|
- Location F1: 0.7500 |
|
- Person F1: 0.7733 |
|
- Product F1: 0.4152 |
|
|
|
## 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: 16 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 100 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Overall Precision | Overall Recall | Overall F1 | Overall Accuracy | Corporation F1 | Creative-work F1 | Group F1 | Location F1 | Person F1 | Product F1 | |
|
|:-------------:|:-----:|:----:|:---------------:|:-----------------:|:--------------:|:----------:|:----------------:|:--------------:|:----------------:|:--------:|:-----------:|:---------:|:----------:| |
|
| 0.2691 | 1.0 | 213 | 0.4035 | 0.0 | 0.0 | 0.0 | 0.8979 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | 0.0 | |
|
| 0.1604 | 2.0 | 426 | 0.3054 | 0.6255 | 0.4161 | 0.4998 | 0.9324 | 0.0 | 0.0 | 0.0 | 0.3534 | 0.6877 | 0.0 | |
|
| 0.1118 | 3.0 | 639 | 0.2864 | 0.6655 | 0.4643 | 0.5470 | 0.9404 | 0.1961 | 0.1164 | 0.1538 | 0.5803 | 0.7221 | 0.1865 | |
|
| 0.0524 | 4.0 | 852 | 0.2891 | 0.6945 | 0.5042 | 0.5842 | 0.9442 | 0.2017 | 0.3273 | 0.2472 | 0.6522 | 0.7366 | 0.2581 | |
|
| 0.0446 | 5.0 | 1065 | 0.2691 | 0.6815 | 0.5847 | 0.6294 | 0.9486 | 0.2737 | 0.3415 | 0.3007 | 0.6703 | 0.7768 | 0.3243 | |
|
| 0.0296 | 6.0 | 1278 | 0.2739 | 0.6740 | 0.5615 | 0.6126 | 0.9479 | 0.3065 | 0.3766 | 0.3333 | 0.7 | 0.7582 | 0.3472 | |
|
| 0.0261 | 7.0 | 1491 | 0.3150 | 0.6907 | 0.5415 | 0.6071 | 0.9457 | 0.2292 | 0.3350 | 0.304 | 0.6369 | 0.7547 | 0.2982 | |
|
| 0.0193 | 8.0 | 1704 | 0.2922 | 0.6957 | 0.5772 | 0.6310 | 0.9496 | 0.2887 | 0.3621 | 0.3676 | 0.7475 | 0.7645 | 0.4158 | |
|
| 0.0173 | 9.0 | 1917 | 0.2823 | 0.6845 | 0.5963 | 0.6374 | 0.9501 | 0.25 | 0.3863 | 0.3660 | 0.6729 | 0.7810 | 0.4064 | |
|
| 0.0227 | 10.0 | 2130 | 0.2912 | 0.6719 | 0.5681 | 0.6157 | 0.9482 | 0.2268 | 0.3797 | 0.3625 | 0.7045 | 0.7572 | 0.4286 | |
|
| 0.0185 | 11.0 | 2343 | 0.3140 | 0.6941 | 0.5598 | 0.6198 | 0.9482 | 0.2532 | 0.3896 | 0.3382 | 0.7059 | 0.7601 | 0.3961 | |
|
| 0.0221 | 12.0 | 2556 | 0.3527 | 0.6937 | 0.5473 | 0.6119 | 0.9470 | 0.3220 | 0.3687 | 0.35 | 0.7245 | 0.7502 | 0.3308 | |
|
| 0.0099 | 13.0 | 2769 | 0.3332 | 0.6872 | 0.5748 | 0.6260 | 0.9493 | 0.3168 | 0.3782 | 0.3597 | 0.7391 | 0.7627 | 0.4027 | |
|
| 0.0062 | 14.0 | 2982 | 0.3637 | 0.7287 | 0.5465 | 0.6246 | 0.9479 | 0.25 | 0.3700 | 0.4065 | 0.7340 | 0.7526 | 0.3468 | |
|
| 0.0075 | 15.0 | 3195 | 0.3239 | 0.6913 | 0.5914 | 0.6374 | 0.9499 | 0.2703 | 0.3636 | 0.4030 | 0.7500 | 0.7733 | 0.4152 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.17.0 |
|
- Pytorch 1.11.0+cu113 |
|
- Datasets 2.0.0 |
|
- Tokenizers 0.11.6 |
|
|
