|
--- |
|
datasets: |
|
- amazon_reviews_multi |
|
language: |
|
- en |
|
library_name: transformers |
|
tags: |
|
- Text Classification |
|
- Pytorch |
|
- Sentiment_Analysis |
|
- Deberta |
|
license: mit |
|
--- |
|
# Deberta for Sentiment Analysis |
|
|
|
This is a Deberta model finetuned on over 1 million reviews from Amazon's multi-reviews dataset. |
|
|
|
|
|
## How to use the model |
|
|
|
```python |
|
import torch |
|
from transformers import AutoModelForSequenceClassification, AutoTokenizer |
|
|
|
def get_sentiment(sentence): |
|
bert_dict = {} |
|
vectors = tokenizer(sentence, return_tensors='pt').to(device) |
|
outputs = bert_model(**vectors).logits |
|
probs = torch.nn.functional.softmax(outputs, dim = 1)[0] |
|
bert_dict['neg'] = round(probs[0].item(), 3) |
|
bert_dict['neu'] = round(probs[1].item(), 3) |
|
bert_dict['pos'] = round(probs[2].item(), 3) |
|
return bert_dict |
|
|
|
MODEL_NAME = 'RashidNLP/Amazon-Deberta-Base-Sentiment' |
|
device = torch.device('cuda' if torch.cuda.is_available() else 'cpu') |
|
|
|
bert_model = AutoModelForSequenceClassification.from_pretrained(MODEL_NAME, num_labels = 3).to(device) |
|
tokenizer = AutoTokenizer.from_pretrained(MODEL_NAME) |
|
|
|
get_sentiment("This is quite a mess you have made") |
|
|
|
``` |