|
--- |
|
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-04 |
|
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-04 |
|
|
|
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: 0.0382 |
|
- Accuracy: 0.9878 |
|
- F1: 0.6398 |
|
- Precision: 0.7120 |
|
- Recall: 0.5810 |
|
|
|
## 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 | |
|
|:-------------:|:------:|:-----:|:---------------:|:--------:|:------:|:---------:|:------:| |
|
| 0.0934 | 0.4929 | 1000 | 0.0902 | 0.9814 | 0.0 | 0.0 | 0.0 | |
|
| 0.0778 | 0.9857 | 2000 | 0.0701 | 0.9814 | 0.0 | 0.0 | 0.0 | |
|
| 0.0618 | 1.4786 | 3000 | 0.0565 | 0.9828 | 0.1749 | 0.8221 | 0.0978 | |
|
| 0.0535 | 1.9714 | 4000 | 0.0488 | 0.9842 | 0.3301 | 0.7895 | 0.2087 | |
|
| 0.0473 | 2.4643 | 5000 | 0.0452 | 0.9856 | 0.4668 | 0.7510 | 0.3386 | |
|
| 0.0436 | 2.9571 | 6000 | 0.0424 | 0.9860 | 0.4963 | 0.7467 | 0.3717 | |
|
| 0.0389 | 3.4500 | 7000 | 0.0403 | 0.9865 | 0.5326 | 0.7503 | 0.4128 | |
|
| 0.0376 | 3.9428 | 8000 | 0.0396 | 0.9865 | 0.5587 | 0.7128 | 0.4594 | |
|
| 0.0339 | 4.4357 | 9000 | 0.0388 | 0.9867 | 0.5583 | 0.7351 | 0.4500 | |
|
| 0.0337 | 4.9285 | 10000 | 0.0385 | 0.9871 | 0.5737 | 0.7467 | 0.4658 | |
|
| 0.0295 | 5.4214 | 11000 | 0.0377 | 0.9871 | 0.6013 | 0.7109 | 0.5210 | |
|
| 0.0305 | 5.9142 | 12000 | 0.0383 | 0.9871 | 0.5951 | 0.7187 | 0.5078 | |
|
| 0.0254 | 6.4071 | 13000 | 0.0373 | 0.9874 | 0.6115 | 0.7197 | 0.5316 | |
|
| 0.0273 | 6.9000 | 14000 | 0.0378 | 0.9876 | 0.6175 | 0.7268 | 0.5367 | |
|
| 0.0228 | 7.3928 | 15000 | 0.0379 | 0.9875 | 0.6101 | 0.7257 | 0.5262 | |
|
| 0.0235 | 7.8857 | 16000 | 0.0380 | 0.9872 | 0.6269 | 0.6861 | 0.5772 | |
|
| 0.0208 | 8.3785 | 17000 | 0.0382 | 0.9877 | 0.6348 | 0.7077 | 0.5756 | |
|
| 0.0204 | 8.8714 | 18000 | 0.0382 | 0.9878 | 0.6398 | 0.7120 | 0.5810 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.2 |
|
- Pytorch 2.4.1+cu121 |
|
- Datasets 2.21.0 |
|
- Tokenizers 0.19.1 |
|
|
