|
--- |
|
library_name: transformers |
|
license: mit |
|
base_model: microsoft/deberta-v3-small |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- f1 |
|
- precision |
|
- recall |
|
model-index: |
|
- name: doc-topic-model_eval-01_train-03 |
|
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. --> |
|
|
|
# doc-topic-model_eval-01_train-03 |
|
|
|
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: 0.0378 |
|
- Accuracy: 0.9879 |
|
- F1: 0.6261 |
|
- Precision: 0.7349 |
|
- Recall: 0.5454 |
|
|
|
## 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: 4 |
|
- eval_batch_size: 256 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 100 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | F1 | Precision | Recall | |
|
|:-------------:|:------:|:-----:|:---------------:|:--------:|:------:|:---------:|:------:| |
|
| 0.0935 | 0.4931 | 1000 | 0.0898 | 0.9814 | 0.0 | 0.0 | 0.0 | |
|
| 0.0764 | 0.9862 | 2000 | 0.0702 | 0.9814 | 0.0 | 0.0 | 0.0 | |
|
| 0.0621 | 1.4793 | 3000 | 0.0570 | 0.9820 | 0.0695 | 0.8912 | 0.0362 | |
|
| 0.0542 | 1.9724 | 4000 | 0.0498 | 0.9840 | 0.2864 | 0.8319 | 0.1730 | |
|
| 0.0468 | 2.4655 | 5000 | 0.0468 | 0.9852 | 0.4191 | 0.7753 | 0.2872 | |
|
| 0.0441 | 2.9586 | 6000 | 0.0435 | 0.9861 | 0.4898 | 0.7741 | 0.3582 | |
|
| 0.0395 | 3.4517 | 7000 | 0.0418 | 0.9860 | 0.5279 | 0.7116 | 0.4196 | |
|
| 0.0384 | 3.9448 | 8000 | 0.0401 | 0.9866 | 0.5588 | 0.7206 | 0.4564 | |
|
| 0.0343 | 4.4379 | 9000 | 0.0392 | 0.9869 | 0.5774 | 0.7226 | 0.4809 | |
|
| 0.0337 | 4.9310 | 10000 | 0.0378 | 0.9873 | 0.5919 | 0.7400 | 0.4932 | |
|
| 0.0305 | 5.4241 | 11000 | 0.0373 | 0.9876 | 0.5989 | 0.7503 | 0.4983 | |
|
| 0.0295 | 5.9172 | 12000 | 0.0378 | 0.9875 | 0.6108 | 0.7303 | 0.5249 | |
|
| 0.0271 | 6.4103 | 13000 | 0.0375 | 0.9877 | 0.6080 | 0.7490 | 0.5116 | |
|
| 0.0257 | 6.9034 | 14000 | 0.0377 | 0.9876 | 0.6145 | 0.7284 | 0.5313 | |
|
| 0.0234 | 7.3964 | 15000 | 0.0377 | 0.9876 | 0.6243 | 0.7147 | 0.5542 | |
|
| 0.0241 | 7.8895 | 16000 | 0.0378 | 0.9879 | 0.6261 | 0.7349 | 0.5454 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.2 |
|
- Pytorch 2.4.1+cu121 |
|
- Datasets 2.21.0 |
|
- Tokenizers 0.19.1 |
|
|
