|
--- |
|
license: apache-2.0 |
|
base_model: riotu-lab/ArabianGPT-01B |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- bleu |
|
- rouge |
|
model-index: |
|
- name: nw_egy |
|
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. --> |
|
|
|
# nw_egy |
|
|
|
This model is a fine-tuned version of [riotu-lab/ArabianGPT-01B](https://huggingface.co/riotu-lab/ArabianGPT-01B) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 2.5201 |
|
- Bleu: 0.2511 |
|
- Rouge1: 0.4534 |
|
- Rouge2: 0.2137 |
|
- Rougel: 0.3984 |
|
|
|
## 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: 5e-05 |
|
- train_batch_size: 8 |
|
- eval_batch_size: 8 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 500 |
|
- num_epochs: 3.0 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Bleu | Rouge1 | Rouge2 | Rougel | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:------:|:------:| |
|
| 4.0723 | 1.0 | 442 | 2.9581 | 0.2228 | 0.3913 | 0.1648 | 0.3201 | |
|
| 2.8055 | 2.0 | 884 | 2.6063 | 0.2415 | 0.4347 | 0.1971 | 0.3777 | |
|
| 2.5505 | 3.0 | 1326 | 2.5201 | 0.2511 | 0.4534 | 0.2137 | 0.3984 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.45.0.dev0 |
|
- Pytorch 2.3.1+cu121 |
|
- Datasets 2.19.2 |
|
- Tokenizers 0.19.1 |
|
|
