|
|
--- |
|
|
library_name: transformers |
|
|
license: mit |
|
|
base_model: intfloat/e5-small |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- accuracy |
|
|
- precision |
|
|
- recall |
|
|
- f1 |
|
|
model-index: |
|
|
- name: intfloat-e5-small-arabic-fp16 |
|
|
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. --> |
|
|
|
|
|
# intfloat-e5-small-arabic-fp16 |
|
|
|
|
|
This model is a fine-tuned version of [intfloat/e5-small](https://huggingface.co/intfloat/e5-small) on an unknown dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.7833 |
|
|
- Accuracy: 0.6782 |
|
|
- Precision: 0.6650 |
|
|
- Recall: 0.6782 |
|
|
- F1: 0.6560 |
|
|
|
|
|
## 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: 64 |
|
|
- eval_batch_size: 64 |
|
|
- seed: 42 |
|
|
- gradient_accumulation_steps: 2 |
|
|
- total_train_batch_size: 128 |
|
|
- optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments |
|
|
- lr_scheduler_type: linear |
|
|
- lr_scheduler_warmup_ratio: 0.3 |
|
|
- num_epochs: 10 |
|
|
- mixed_precision_training: Native AMP |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Accuracy | Precision | Recall | F1 | |
|
|
|:-------------:|:------:|:----:|:---------------:|:--------:|:---------:|:------:|:------:| |
|
|
| 1.0984 | 0.3636 | 50 | 1.0847 | 0.4373 | 0.5309 | 0.4373 | 0.2662 | |
|
|
| 1.069 | 0.7273 | 100 | 1.0270 | 0.5677 | 0.6816 | 0.5677 | 0.4864 | |
|
|
| 1.0003 | 1.0873 | 150 | 0.9488 | 0.6041 | 0.6923 | 0.6041 | 0.5281 | |
|
|
| 0.9438 | 1.4509 | 200 | 0.9085 | 0.6095 | 0.7017 | 0.6095 | 0.5340 | |
|
|
| 0.9178 | 1.8145 | 250 | 0.9204 | 0.5895 | 0.6283 | 0.5895 | 0.5182 | |
|
|
| 0.8936 | 2.1745 | 300 | 0.8394 | 0.6441 | 0.5970 | 0.6441 | 0.5653 | |
|
|
| 0.8824 | 2.5382 | 350 | 0.8447 | 0.6464 | 0.7275 | 0.6464 | 0.5674 | |
|
|
| 0.8774 | 2.9018 | 400 | 0.8706 | 0.625 | 0.6165 | 0.625 | 0.5859 | |
|
|
| 0.8655 | 3.2618 | 450 | 0.8125 | 0.6541 | 0.6337 | 0.6541 | 0.6378 | |
|
|
| 0.8517 | 3.6255 | 500 | 0.8528 | 0.6477 | 0.6565 | 0.6477 | 0.6013 | |
|
|
| 0.8451 | 3.9891 | 550 | 0.7914 | 0.6686 | 0.6502 | 0.6686 | 0.6340 | |
|
|
| 0.8084 | 4.3491 | 600 | 0.7833 | 0.6782 | 0.6650 | 0.6782 | 0.6560 | |
|
|
| 0.8028 | 4.7127 | 650 | 0.7635 | 0.6923 | 0.6832 | 0.6923 | 0.6803 | |
|
|
| 0.7874 | 5.0727 | 700 | 0.7815 | 0.6709 | 0.6828 | 0.6709 | 0.6754 | |
|
|
| 0.7861 | 5.4364 | 750 | 0.7629 | 0.6873 | 0.6811 | 0.6873 | 0.6820 | |
|
|
| 0.7593 | 5.8 | 800 | 0.7794 | 0.6823 | 0.6744 | 0.6823 | 0.6684 | |
|
|
| 0.753 | 6.16 | 850 | 0.7680 | 0.6845 | 0.6859 | 0.6845 | 0.6838 | |
|
|
| 0.753 | 6.5236 | 900 | 0.8404 | 0.6395 | 0.6783 | 0.6395 | 0.6459 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.51.1 |
|
|
- Pytorch 2.6.0+cu124 |
|
|
- Datasets 3.5.0 |
|
|
- Tokenizers 0.21.1 |
|
|
|
