|
|
--- |
|
|
language: en |
|
|
license: apache-2.0 |
|
|
tags: |
|
|
- sentiment analysis |
|
|
- text classification |
|
|
- bert |
|
|
- transformers |
|
|
- news |
|
|
- reviews |
|
|
--- |
|
|
|
|
|
# SentimentBERT — Fine-tuned BERT for Sentiment Classification (Positive, Neutral, Negative) |
|
|
|
|
|
**SentimentBERT** is a Finetuned BERT-based model specifically for **sentiment classification of sentences** into three categories: **Positive**, **Negative**, and **Neutral**. |
|
|
|
|
|
This model has been trained on a ** 130K large and diverse dataset of news articles** across a wide range of categories. It achieves **over 86% accuracy** and demonstrates a strong understanding of sentence-level sentiment, even in nuanced or mixed-context cases. |
|
|
|
|
|
--- |
|
|
|
|
|
## Model Highlights |
|
|
|
|
|
- **Base model**: `bert-base-uncased` |
|
|
- **Fine tuned for**: Sentiment classification (3-class) |
|
|
- **Accuracy**: > 86% |
|
|
- **Classes**: Positive, Neutral, Negative |
|
|
- **Language**: English |
|
|
- **Format**: `safetensors` |
|
|
- **Tokenizer**: Compatible with `bert-base-uncased` |
|
|
|
|
|
--- |
|
|
|
|
|
## Applications |
|
|
|
|
|
This model is well-suited for: |
|
|
|
|
|
- **News article sentiment analysis** |
|
|
- **Amazon product review analysis** |
|
|
- **Customer support or service feedback systems** |
|
|
- **General-purpose opinion mining** |
|
|
|
|
|
|
|
|
|
|
|
Thanks for visiting and downloading this model! |
|
|
If this model helped you, please consider leaving a like. Your support helps this model reach more developers and encourages further improvements if any. |
|
|
--- |
|
|
|
|
|
## How to use the model |
|
|
|
|
|
```python |
|
|
from transformers import AutoTokenizer, AutoModelForSequenceClassification |
|
|
import torch |
|
|
|
|
|
model = AutoModelForSequenceClassification.from_pretrained("mervp/SentimentBERT") |
|
|
tokenizer = AutoTokenizer.from_pretrained("mervp/SentimentBERT") |
|
|
|
|
|
def predict_sentiment(text): |
|
|
model.eval() |
|
|
inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True) |
|
|
with torch.no_grad(): |
|
|
outputs = model(**inputs) |
|
|
logits = outputs.logits |
|
|
prediction = torch.argmax(logits, dim=-1).item() |
|
|
label = model.config.id2label[prediction] |
|
|
return label |
|
|
|
|
|
print(predict_sentiment("What a beautiful day.")) # positive |
|
|
print(predict_sentiment("The service was excellent.")) # positive |
|
|
print(predict_sentiment("He did a fantastic job.")) # positive |
|
|
print(predict_sentiment("The experience was terrible.")) # negative |
|
|
print(predict_sentiment("Everything went wrong.")) # negative |
|
|
print(predict_sentiment("He opened the door and walked in.")) # neutral |
|
|
print(predict_sentiment("They are meeting at 5 PM.")) # neutral |
|
|
print(predict_sentiment("She has a cat.")) # neutral |
|
|
|
