|
--- |
|
library_name: transformers |
|
base_model: microsoft/deberta-v3-small |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- accuracy |
|
- f1 |
|
- precision |
|
- recall |
|
model-index: |
|
- name: doc-topic-model_eval-00_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-00_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: 0.0375 |
|
- Accuracy: 0.9878 |
|
- F1: 0.6375 |
|
- Precision: 0.7079 |
|
- Recall: 0.5798 |
|
|
|
## 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 | |
|
|:-------------:|:------:|:-----:|:---------------:|:--------:|:------:|:---------:|:------:| |
|
| 0.0923 | 0.4931 | 1000 | 0.0865 | 0.9815 | 0.0 | 0.0 | 0.0 | |
|
| 0.0734 | 0.9862 | 2000 | 0.0668 | 0.9815 | 0.0 | 0.0 | 0.0 | |
|
| 0.0606 | 1.4793 | 3000 | 0.0552 | 0.9824 | 0.1265 | 0.7701 | 0.0689 | |
|
| 0.0532 | 1.9724 | 4000 | 0.0491 | 0.9841 | 0.2944 | 0.8178 | 0.1795 | |
|
| 0.0466 | 2.4655 | 5000 | 0.0467 | 0.9851 | 0.4323 | 0.7342 | 0.3063 | |
|
| 0.0433 | 2.9586 | 6000 | 0.0428 | 0.9859 | 0.4847 | 0.7565 | 0.3566 | |
|
| 0.0391 | 3.4517 | 7000 | 0.0408 | 0.9866 | 0.5389 | 0.7450 | 0.4221 | |
|
| 0.0378 | 3.9448 | 8000 | 0.0395 | 0.9867 | 0.5527 | 0.7365 | 0.4423 | |
|
| 0.0338 | 4.4379 | 9000 | 0.0387 | 0.9870 | 0.5844 | 0.7160 | 0.4936 | |
|
| 0.0333 | 4.9310 | 10000 | 0.0380 | 0.9871 | 0.5953 | 0.7094 | 0.5128 | |
|
| 0.0301 | 5.4241 | 11000 | 0.0371 | 0.9876 | 0.6042 | 0.7368 | 0.5120 | |
|
| 0.0292 | 5.9172 | 12000 | 0.0367 | 0.9877 | 0.6120 | 0.7381 | 0.5227 | |
|
| 0.0266 | 6.4103 | 13000 | 0.0369 | 0.9879 | 0.6132 | 0.7535 | 0.5170 | |
|
| 0.0257 | 6.9034 | 14000 | 0.0371 | 0.9877 | 0.6187 | 0.7257 | 0.5392 | |
|
| 0.0229 | 7.3964 | 15000 | 0.0371 | 0.9880 | 0.6330 | 0.7279 | 0.5600 | |
|
| 0.0238 | 7.8895 | 16000 | 0.0372 | 0.9879 | 0.6306 | 0.7250 | 0.5579 | |
|
| 0.0198 | 8.3826 | 17000 | 0.0375 | 0.9878 | 0.6375 | 0.7079 | 0.5798 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.2 |
|
- Pytorch 2.4.1+cu121 |
|
- Datasets 2.21.0 |
|
- Tokenizers 0.19.1 |
|
|
