|
|
--- |
|
|
license: apache-2.0 |
|
|
tags: |
|
|
- generated_from_trainer |
|
|
metrics: |
|
|
- bleu |
|
|
model-index: |
|
|
- name: mt5-aym-zero |
|
|
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. --> |
|
|
|
|
|
# mt5-aym-zero |
|
|
|
|
|
This model is a fine-tuned version of [alvations/mt5-aym-zero](https://huggingface.co/alvations/mt5-aym-zero) on the None dataset. |
|
|
It achieves the following results on the evaluation set: |
|
|
- Loss: 0.2035 |
|
|
- Chrf: 23.9937 |
|
|
- Bleu: 2.9371 |
|
|
- Gen Len: 17.1956 |
|
|
|
|
|
## 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 |
|
|
- training_steps: 100000 |
|
|
|
|
|
### Training results |
|
|
|
|
|
| Training Loss | Epoch | Step | Bleu | Chrf | Gen Len | Validation Loss | |
|
|
|:-------------:|:-----:|:------:|:------:|:-------:|:-------:|:---------------:| |
|
|
| 0.0658 | 5.13 | 20000 | 2.9963 | 23.9471 | 17.2839 | 0.1892 | |
|
|
| 0.0633 | 10.26 | 40000 | 2.9953 | 24.004 | 17.2648 | 0.1932 | |
|
|
| 0.059 | 15.39 | 60000 | 2.9368 | 24.0104 | 17.2207 | 0.1975 | |
|
|
| 0.057 | 20.52 | 80000 | 2.9962 | 23.978 | 17.1725 | 0.2006 | |
|
|
| 0.055 | 25.65 | 100000 | 0.2035 | 23.9937 | 2.9371 | 17.1956 | |
|
|
|
|
|
|
|
|
### Framework versions |
|
|
|
|
|
- Transformers 4.29.2 |
|
|
- Pytorch 2.0.0+cu118 |
|
|
- Datasets 2.12.0 |
|
|
- Tokenizers 0.13.3 |
|
|
|
