|
|
--- |
|
|
license: apache-2.0 |
|
|
library_name: peft |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
base_model: google/madlad400-3b-mt |
|
|
datasets: |
|
|
- generator |
|
|
metrics: |
|
|
- bleu |
|
|
model-index: |
|
|
- name: traduction_fon_french |
|
|
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. --> |
|
|
|
|
|
# traduction_fon_french |
|
|
|
|
|
This model is a fine-tuned version of [google/madlad400-3b-mt](https://huggingface.co/google/madlad400-3b-mt) on the generator dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 4.1804 |
|
|
- Bleu: 4.0833 |
|
|
- Gen Len: 7.1314 |
|
|
|
|
|
## 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: 0.00016 |
|
|
- train_batch_size: 18 |
|
|
- eval_batch_size: 18 |
|
|
- seed: 42 |
|
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
|
- lr_scheduler_type: linear |
|
|
- num_epochs: 3 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Bleu | Gen Len | |
|
|
|:-------------:|:-----:|:-----:|:---------------:|:------:|:-------:| |
|
|
| 2.1185 | 1.0 | 4957 | 4.1772 | 3.9348 | 7.1858 | |
|
|
| 2.0246 | 2.0 | 9914 | 4.1778 | 4.1343 | 7.1091 | |
|
|
| 1.9755 | 3.0 | 14871 | 4.1804 | 4.0833 | 7.1314 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- PEFT 0.11.1 |
|
|
- Transformers 4.41.2 |
|
|
- Pytorch 2.1.2 |
|
|
- Datasets 2.19.2 |
|
|
- Tokenizers 0.19.1 |
