|
--- |
|
license: mit |
|
base_model: microsoft/deberta-v3-small |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- f1 |
|
- precision |
|
- recall |
|
model-index: |
|
- name: legistorm-categorizer-seqclass-deberta-v1 |
|
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. --> |
|
|
|
# legistorm-categorizer-seqclass-deberta-v1 |
|
|
|
This model is a fine-tuned version of [microsoft/deberta-v3-small](https://huggingface.co/microsoft/deberta-v3-small) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.0265 |
|
- Accuracy: 0.9954 |
|
- F1: 0.9623 |
|
- Precision: 0.9814 |
|
- Recall: 0.9439 |
|
|
|
## 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: 24 |
|
- eval_batch_size: 24 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 6 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | |
|
|:-------------:|:-----:|:-----:|:---------------:|:--------:|:------:|:---------:|:------:| |
|
| No log | 0.25 | 456 | 0.0191 | 0.9957 | 0.9644 | 0.9973 | 0.9337 | |
|
| 0.0725 | 0.5 | 912 | 0.0176 | 0.9960 | 0.9671 | 1.0 | 0.9362 | |
|
| 0.0044 | 0.75 | 1368 | 0.0161 | 0.9963 | 0.9699 | 0.9973 | 0.9439 | |
|
| 0.0015 | 1.0 | 1824 | 0.0183 | 0.9963 | 0.9699 | 0.9973 | 0.9439 | |
|
| 0.0007 | 1.25 | 2280 | 0.0189 | 0.9962 | 0.9686 | 0.9946 | 0.9439 | |
|
| 0.0004 | 1.5 | 2736 | 0.0197 | 0.9962 | 0.9686 | 0.9946 | 0.9439 | |
|
| 0.0003 | 1.75 | 3192 | 0.0181 | 0.9963 | 0.9699 | 0.9946 | 0.9464 | |
|
| 0.0002 | 2.0 | 3648 | 0.0210 | 0.9957 | 0.9647 | 0.9893 | 0.9413 | |
|
| 0.0001 | 2.25 | 4104 | 0.0224 | 0.9959 | 0.9661 | 0.9893 | 0.9439 | |
|
| 0.0001 | 2.5 | 4560 | 0.0228 | 0.9959 | 0.9661 | 0.9893 | 0.9439 | |
|
| 0.0001 | 2.75 | 5016 | 0.0231 | 0.9957 | 0.9648 | 0.9867 | 0.9439 | |
|
| 0.0001 | 3.0 | 5472 | 0.0224 | 0.9957 | 0.9648 | 0.9867 | 0.9439 | |
|
| 0.0001 | 3.25 | 5928 | 0.0218 | 0.9959 | 0.9661 | 0.9893 | 0.9439 | |
|
| 0.0 | 3.5 | 6384 | 0.0232 | 0.9955 | 0.9635 | 0.9840 | 0.9439 | |
|
| 0.0 | 3.75 | 6840 | 0.0229 | 0.9957 | 0.9649 | 0.9841 | 0.9464 | |
|
| 0.0 | 4.0 | 7296 | 0.0240 | 0.9955 | 0.9635 | 0.9840 | 0.9439 | |
|
| 0.0 | 4.25 | 7752 | 0.0249 | 0.9957 | 0.9648 | 0.9867 | 0.9439 | |
|
| 0.0 | 4.5 | 8208 | 0.0251 | 0.9954 | 0.9623 | 0.9814 | 0.9439 | |
|
| 0.0 | 4.74 | 8664 | 0.0248 | 0.9954 | 0.9623 | 0.9814 | 0.9439 | |
|
| 0.0 | 4.99 | 9120 | 0.0251 | 0.9954 | 0.9623 | 0.9814 | 0.9439 | |
|
| 0.0 | 5.24 | 9576 | 0.0253 | 0.9954 | 0.9623 | 0.9814 | 0.9439 | |
|
| 0.0 | 5.49 | 10032 | 0.0266 | 0.9954 | 0.9623 | 0.9814 | 0.9439 | |
|
| 0.0 | 5.74 | 10488 | 0.0264 | 0.9954 | 0.9623 | 0.9814 | 0.9439 | |
|
| 0.0 | 5.99 | 10944 | 0.0265 | 0.9954 | 0.9623 | 0.9814 | 0.9439 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.36.2 |
|
- Pytorch 2.1.2+cu121 |
|
- Datasets 2.16.1 |
|
- Tokenizers 0.15.0 |
|
|
