|
--- |
|
library_name: transformers |
|
license: mit |
|
base_model: xlnet/xlnet-base-cased |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
model-index: |
|
- name: xlnet-base-cased-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. --> |
|
|
|
# xlnet-base-cased-grammar-ner |
|
|
|
This model is a fine-tuned version of [xlnet/xlnet-base-cased](https://huggingface.co/xlnet/xlnet-base-cased) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.1027 |
|
- Accuracy: 0.9879 |
|
- F1 Macro: 0.7899 |
|
- F1 Micro: 0.9048 |
|
- Precision Macro: 0.8321 |
|
- Precision Micro: 0.9436 |
|
- Recall Macro: 0.7769 |
|
- Recall Micro: 0.8691 |
|
|
|
## 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 | |
|
|:-------------:|:-----:|:----:|:---------------:|:--------:|:--------:|:--------:|:---------------:|:---------------:|:------------:|:------------:| |
|
| 0.3654 | 1.0 | 93 | 0.2315 | 0.9329 | 0.2241 | 0.4091 | 0.2338 | 0.5016 | 0.2436 | 0.3454 | |
|
| 0.1834 | 2.0 | 186 | 0.1603 | 0.9627 | 0.2982 | 0.6748 | 0.4007 | 0.8441 | 0.2733 | 0.5621 | |
|
| 0.1161 | 3.0 | 279 | 0.1472 | 0.9649 | 0.4016 | 0.7249 | 0.4585 | 0.7494 | 0.3910 | 0.7020 | |
|
| 0.0764 | 4.0 | 372 | 0.1233 | 0.9739 | 0.4844 | 0.7783 | 0.5824 | 0.8346 | 0.4487 | 0.7291 | |
|
| 0.0561 | 5.0 | 465 | 0.1224 | 0.9738 | 0.4467 | 0.7792 | 0.5923 | 0.8650 | 0.4044 | 0.7088 | |
|
| 0.0423 | 6.0 | 558 | 0.1135 | 0.9799 | 0.6021 | 0.8375 | 0.6885 | 0.8825 | 0.5619 | 0.7968 | |
|
| 0.0319 | 7.0 | 651 | 0.0987 | 0.9820 | 0.6386 | 0.8541 | 0.6928 | 0.8841 | 0.6342 | 0.8262 | |
|
| 0.0221 | 8.0 | 744 | 0.1034 | 0.9836 | 0.6463 | 0.8623 | 0.7605 | 0.9184 | 0.6045 | 0.8126 | |
|
| 0.0175 | 9.0 | 837 | 0.0984 | 0.9852 | 0.6852 | 0.8794 | 0.7154 | 0.9045 | 0.6849 | 0.8555 | |
|
| 0.0094 | 10.0 | 930 | 0.0985 | 0.9865 | 0.6985 | 0.8936 | 0.7434 | 0.9272 | 0.6914 | 0.8623 | |
|
| 0.0078 | 11.0 | 1023 | 0.0987 | 0.9858 | 0.7314 | 0.8876 | 0.7563 | 0.9119 | 0.7406 | 0.8646 | |
|
| 0.0056 | 12.0 | 1116 | 0.1051 | 0.9868 | 0.7501 | 0.8923 | 0.7955 | 0.9270 | 0.7410 | 0.8600 | |
|
| 0.0047 | 13.0 | 1209 | 0.1027 | 0.9866 | 0.7606 | 0.8936 | 0.8116 | 0.9272 | 0.7469 | 0.8623 | |
|
| 0.0031 | 14.0 | 1302 | 0.1009 | 0.9866 | 0.7762 | 0.8953 | 0.8034 | 0.9233 | 0.7769 | 0.8691 | |
|
| 0.0027 | 15.0 | 1395 | 0.1008 | 0.9873 | 0.7763 | 0.8995 | 0.8073 | 0.9322 | 0.7769 | 0.8691 | |
|
| 0.002 | 16.0 | 1488 | 0.1034 | 0.9884 | 0.7939 | 0.9067 | 0.8590 | 0.9505 | 0.7591 | 0.8668 | |
|
| 0.0015 | 17.0 | 1581 | 0.1020 | 0.9881 | 0.7925 | 0.9059 | 0.8362 | 0.9459 | 0.7769 | 0.8691 | |
|
| 0.0016 | 18.0 | 1674 | 0.1027 | 0.9879 | 0.7899 | 0.9048 | 0.8321 | 0.9436 | 0.7769 | 0.8691 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.46.3 |
|
- Pytorch 2.5.1+cu121 |
|
- Datasets 3.2.0 |
|
- Tokenizers 0.20.3 |
|
|