|
--- |
|
tags: |
|
- generated_from_trainer |
|
model-index: |
|
- name: uribe-twitter-assistant-30ep |
|
results: [] |
|
widget: |
|
- text: Santos y la Paz |
|
- text: Las Farc son |
|
datasets: |
|
- jhonparra18/alvarouribevel-tweets |
|
language: |
|
- es |
|
--- |
|
|
|
<!-- 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. --> |
|
|
|
# uribe-twitter-assistant-30ep |
|
|
|
This model is a fine-tuned version of [flax-community/gpt-2-spanish](https://huggingface.co/flax-community/gpt-2-spanish) on an unknown dataset. |
|
It achieves the following results on the evaluation set: |
|
- Loss: 3.3026 |
|
|
|
## 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: 0.0001 |
|
- train_batch_size: 22 |
|
- eval_batch_size: 22 |
|
- seed: 42 |
|
- gradient_accumulation_steps: 2 |
|
- total_train_batch_size: 44 |
|
- optimizer: Adam with betas=(0.9,0.999) and epsilon=1e-08 |
|
- lr_scheduler_type: cosine |
|
- lr_scheduler_warmup_steps: 100 |
|
- num_epochs: 30 |
|
- mixed_precision_training: Native AMP |
|
|
|
### Training results |
|
|
|
| Training Loss | Epoch | Step | Validation Loss | |
|
|:-------------:|:-----:|:-----:|:---------------:| |
|
| 3.5124 | 0.89 | 1000 | 3.3424 | |
|
| 3.144 | 1.77 | 2000 | 3.2148 | |
|
| 2.9674 | 2.66 | 3000 | 3.1513 | |
|
| 2.8433 | 3.55 | 4000 | 3.1241 | |
|
| 2.7166 | 4.43 | 5000 | 3.1198 | |
|
| 2.6151 | 5.32 | 6000 | 3.1450 | |
|
| 2.5038 | 6.21 | 7000 | 3.1762 | |
|
| 2.4035 | 7.09 | 8000 | 3.2427 | |
|
| 2.2987 | 7.98 | 9000 | 3.2282 | |
|
| 2.1788 | 8.87 | 10000 | 3.3026 | |
|
|
|
|
|
### Framework versions |
|
|
|
- Transformers 4.20.1 |
|
- Pytorch 1.12.0 |
|
- Datasets 2.1.0 |
|
- Tokenizers 0.12.1 |
|
|
