|
--- |
|
library_name: transformers |
|
license: apache-2.0 |
|
base_model: bert-base-uncased |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: bert-base-uncased-grammar-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. --> |
|
|
|
# bert-base-uncased-grammar-ner |
|
|
|
This model is a fine-tuned version of [bert-base-uncased](https://huggingface.co/bert-base-uncased) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1052 |
|
- Accuracy: 0.9895 |
|
- F1 Macro: 0.7899 |
|
- F1 Micro: 0.9212 |
|
- Precision Macro: 0.8429 |
|
- Precision Micro: 0.9694 |
|
- Recall Macro: 0.7572 |
|
- Recall Micro: 0.8776 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 24 |
|
- eval_batch_size: 24 |
|
- seed: 42 |
|
- optimizer: Use adamw_torch with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
- lr_scheduler_type: linear |
|
- num_epochs: 18 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 Macro | F1 Micro | Precision Macro | Precision Micro | Recall Macro | Recall Micro | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:|:---------------:|:---------------:|:------------:|:------------:| |
|
| 0.3673 | 1.0 | 93 | 0.2453 | 0.9284 | 0.1919 | 0.4850 | 0.2701 | 0.4678 | 0.1872 | 0.5035 | |
|
| 0.2176 | 2.0 | 186 | 0.1888 | 0.9439 | 0.2591 | 0.5149 | 0.3936 | 0.6230 | 0.2298 | 0.4388 | |
|
| 0.1418 | 3.0 | 279 | 0.1454 | 0.9666 | 0.3554 | 0.725 | 0.4120 | 0.7902 | 0.3577 | 0.6697 | |
|
| 0.0859 | 4.0 | 372 | 0.1238 | 0.9750 | 0.4365 | 0.7789 | 0.6084 | 0.8540 | 0.3946 | 0.7159 | |
|
| 0.0607 | 5.0 | 465 | 0.1136 | 0.9766 | 0.4979 | 0.7965 | 0.5945 | 0.8606 | 0.4781 | 0.7413 | |
|
| 0.0413 | 6.0 | 558 | 0.1103 | 0.9827 | 0.4995 | 0.8608 | 0.6097 | 0.9629 | 0.4415 | 0.7783 | |
|
| 0.0309 | 7.0 | 651 | 0.1109 | 0.9821 | 0.5654 | 0.8558 | 0.6379 | 0.8842 | 0.5439 | 0.8291 | |
|
| 0.0237 | 8.0 | 744 | 0.1056 | 0.9847 | 0.6330 | 0.8721 | 0.7169 | 0.9227 | 0.5923 | 0.8268 | |
|
| 0.0154 | 9.0 | 837 | 0.1009 | 0.9858 | 0.6639 | 0.8816 | 0.7079 | 0.9352 | 0.6422 | 0.8337 | |
|
| 0.0096 | 10.0 | 930 | 0.1003 | 0.9881 | 0.6783 | 0.9047 | 0.7250 | 0.9470 | 0.6494 | 0.8661 | |
|
| 0.0078 | 11.0 | 1023 | 0.1000 | 0.9889 | 0.7661 | 0.9144 | 0.8075 | 0.9571 | 0.7524 | 0.8753 | |
|
| 0.0052 | 12.0 | 1116 | 0.1046 | 0.9890 | 0.7563 | 0.9166 | 0.7940 | 0.9619 | 0.7561 | 0.8753 | |
|
| 0.0041 | 13.0 | 1209 | 0.1022 | 0.9892 | 0.7804 | 0.9177 | 0.8255 | 0.9644 | 0.7570 | 0.8753 | |
|
| 0.0021 | 14.0 | 1302 | 0.0994 | 0.9887 | 0.7602 | 0.9133 | 0.7959 | 0.9547 | 0.7534 | 0.8753 | |
|
| 0.0018 | 15.0 | 1395 | 0.1043 | 0.9895 | 0.7903 | 0.9212 | 0.8431 | 0.9694 | 0.7572 | 0.8776 | |
|
| 0.0016 | 16.0 | 1488 | 0.1059 | 0.9898 | 0.7901 | 0.9235 | 0.8434 | 0.9744 | 0.7572 | 0.8776 | |
|
| 0.0014 | 17.0 | 1581 | 0.1063 | 0.9898 | 0.7924 | 0.9235 | 0.8472 | 0.9744 | 0.7572 | 0.8776 | |
|
| 0.001 | 18.0 | 1674 | 0.1052 | 0.9895 | 0.7899 | 0.9212 | 0.8429 | 0.9694 | 0.7572 | 0.8776 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.46.3 |
|
- Pytorch 2.5.1+cu121 |
|
- Datasets 3.2.0 |
|
- Tokenizers 0.20.3 |
|
|
