|
--- |
|
inference: false |
|
language: |
|
- bg |
|
license: mit |
|
datasets: |
|
- oscar |
|
- chitanka |
|
- wikipedia |
|
tags: |
|
- torch |
|
--- |
|
|
|
# ROBERTA BASE (cased) trained on private Bulgarian sentiment-analysis dataset |
|
This is a Multilingual Roberta model. |
|
|
|
This model is cased: it does make a difference between bulgarian and Bulgarian. |
|
|
|
### How to use |
|
|
|
Here is how to use this model in PyTorch: |
|
|
|
```python |
|
>>> import torch |
|
>>> from transformers import AutoModel, AutoTokenizer |
|
>>> |
|
>>> model_id = "rmihaylov/roberta-base-sentiment-bg" |
|
>>> model = AutoModel.from_pretrained(model_id, trust_remote_code=True) |
|
>>> tokenizer = AutoTokenizer.from_pretrained(model_id) |
|
>>> |
|
>>> inputs = tokenizer.batch_encode_plus(['Това е умно.', 'Това е тъпо.'], return_tensors='pt') |
|
>>> outputs = model(**inputs) |
|
>>> torch.softmax(outputs, dim=1).tolist() |
|
|
|
[[0.0004746630438603461, 0.9995253086090088], |
|
[0.9986956715583801, 0.0013043134240433574]] |
|
``` |
|
|
