|
--- |
|
license: cc-by-4.0 |
|
base_model: l3cube-pune/tamil-bert |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- precision |
|
- recall |
|
- accuracy |
|
model-index: |
|
- name: Tamil-BERT-finetune-Tamil-questions |
|
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. --> |
|
|
|
# Tamil-BERT-finetune-Tamil-questions |
|
|
|
This model is a fine-tuned version of [l3cube-pune/tamil-bert](https://huggingface.co/l3cube-pune/tamil-bert) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.3564 |
|
- Precision: 0.9226 |
|
- Recall: 0.9218 |
|
- Accuracy: 0.9218 |
|
- F1-score: 0.9220 |
|
|
|
## 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: 16 |
|
- eval_batch_size: 16 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- num_epochs: 8 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | Accuracy | F1-score | |
|
|:-------------:|:-----:|:----:|:---------------:|:---------:|:------:|:--------:|:--------:| |
|
| 1.534 | 1.0 | 305 | 1.2125 | 0.8686 | 0.8778 | 0.8778 | 0.8701 | |
|
| 0.937 | 2.0 | 610 | 0.7374 | 0.8869 | 0.8958 | 0.8958 | 0.8899 | |
|
| 0.5335 | 3.0 | 915 | 0.4742 | 0.8959 | 0.9078 | 0.9078 | 0.9007 | |
|
| 0.3097 | 4.0 | 1220 | 0.3972 | 0.9004 | 0.9138 | 0.9138 | 0.9064 | |
|
| 0.2083 | 5.0 | 1525 | 0.3869 | 0.9103 | 0.9058 | 0.9058 | 0.9018 | |
|
| 0.1535 | 6.0 | 1830 | 0.4181 | 0.9115 | 0.9078 | 0.9078 | 0.9087 | |
|
| 0.1222 | 7.0 | 2135 | 0.3576 | 0.9243 | 0.9238 | 0.9238 | 0.9240 | |
|
| 0.1002 | 8.0 | 2440 | 0.3564 | 0.9226 | 0.9218 | 0.9218 | 0.9220 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.39.3 |
|
- Pytorch 2.1.2 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.15.2 |
|
|
