|
--- |
|
license: apache-2.0 |
|
base_model: dumitrescustefan/mt5-base-romanian |
|
tags: |
|
- generated_from_trainer |
|
metrics: |
|
- bleu |
|
model-index: |
|
- name: ROAI_proba_2 |
|
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. --> |
|
|
|
# ROAI_proba_2 |
|
|
|
This model is a fine-tuned version of [dumitrescustefan/mt5-base-romanian](https://huggingface.co/dumitrescustefan/mt5-base-romanian) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 0.4289 |
|
- Bleu: 0.6870 |
|
- Meteor: 0.8509 |
|
- Rouge-l: 0.8753 |
|
|
|
## 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: 0.0005 |
|
- train_batch_size: 32 |
|
- eval_batch_size: 32 |
|
- seed: 42 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: linear |
|
- lr_scheduler_warmup_steps: 5 |
|
- num_epochs: 3 |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | Bleu | Meteor | Rouge-l | |
|
|:-------------:|:-----:|:----:|:---------------:|:------:|:------:|:-------:| |
|
| 0.6694 | 1.0 | 1207 | 0.5275 | 0.6257 | 0.8122 | 0.8390 | |
|
| 0.8264 | 2.0 | 2414 | 0.4472 | 0.6727 | 0.8430 | 0.8671 | |
|
| 0.5424 | 3.0 | 3621 | 0.4289 | 0.6870 | 0.8509 | 0.8753 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.41.2 |
|
- Pytorch 2.3.0+cu121 |
|
- Datasets 2.18.0 |
|
- Tokenizers 0.19.1 |
|
|
