|
--- |
|
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-02_train-01 |
|
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-02_train-01 |
|
|
|
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: 0.0396 |
|
- Accuracy: 0.9875 |
|
- F1: 0.6321 |
|
- Precision: 0.6977 |
|
- Recall: 0.5777 |
|
|
|
## 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 | |
|
|:-------------:|:------:|:-----:|:---------------:|:--------:|:------:|:---------:|:------:| |
|
| 0.0941 | 0.4931 | 1000 | 0.0907 | 0.9814 | 0.0 | 0.0 | 0.0 | |
|
| 0.0787 | 0.9862 | 2000 | 0.0707 | 0.9814 | 0.0 | 0.0 | 0.0 | |
|
| 0.0628 | 1.4793 | 3000 | 0.0575 | 0.9822 | 0.1225 | 0.7682 | 0.0666 | |
|
| 0.0537 | 1.9724 | 4000 | 0.0503 | 0.9842 | 0.3201 | 0.8086 | 0.1996 | |
|
| 0.0478 | 2.4655 | 5000 | 0.0470 | 0.9851 | 0.4263 | 0.7606 | 0.2961 | |
|
| 0.0453 | 2.9586 | 6000 | 0.0444 | 0.9858 | 0.4983 | 0.7270 | 0.3791 | |
|
| 0.0389 | 3.4517 | 7000 | 0.0419 | 0.9864 | 0.5409 | 0.7312 | 0.4292 | |
|
| 0.0393 | 3.9448 | 8000 | 0.0411 | 0.9863 | 0.5480 | 0.7138 | 0.4447 | |
|
| 0.0349 | 4.4379 | 9000 | 0.0399 | 0.9868 | 0.5747 | 0.7203 | 0.4781 | |
|
| 0.0344 | 4.9310 | 10000 | 0.0391 | 0.9870 | 0.5758 | 0.7380 | 0.4721 | |
|
| 0.0302 | 5.4241 | 11000 | 0.0385 | 0.9871 | 0.5904 | 0.7254 | 0.4977 | |
|
| 0.0305 | 5.9172 | 12000 | 0.0387 | 0.9871 | 0.5966 | 0.7152 | 0.5118 | |
|
| 0.027 | 6.4103 | 13000 | 0.0384 | 0.9874 | 0.6057 | 0.7302 | 0.5174 | |
|
| 0.0282 | 6.9034 | 14000 | 0.0381 | 0.9875 | 0.6079 | 0.7344 | 0.5186 | |
|
| 0.0235 | 7.3964 | 15000 | 0.0385 | 0.9874 | 0.6181 | 0.7103 | 0.5471 | |
|
| 0.0255 | 7.8895 | 16000 | 0.0382 | 0.9876 | 0.6257 | 0.7174 | 0.5548 | |
|
| 0.0214 | 8.3826 | 17000 | 0.0382 | 0.9877 | 0.6353 | 0.7122 | 0.5734 | |
|
| 0.0222 | 8.8757 | 18000 | 0.0388 | 0.9876 | 0.6282 | 0.7127 | 0.5615 | |
|
| 0.0192 | 9.3688 | 19000 | 0.0396 | 0.9875 | 0.6321 | 0.6977 | 0.5777 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.44.2 |
|
- Pytorch 2.4.1+cu121 |
|
- Datasets 2.21.0 |
|
- Tokenizers 0.19.1 |
|
|
