README.md

---
license: mit
datasets:
- mteb/tweet_sentiment_extraction
language:
- en
metrics:
- accuracy
base_model:
- openai-community/gpt2
pipeline_tag: text-classification
library_name: transformers
---
# GPT-2 Sentiment Analysis for Tweets

## Model Details

- **Model Type**: GPT-2 (Fine-tuned for sentiment analysis)
- **Model Architecture**: Transformer-based language model (GPT-2)
- **Fine-tuned On**: `mteb/tweet_sentiment_extraction` dataset
- **Intended Task**: Sentiment Classification (Tweet Sentiment)

## Model Overview

This model is a fine-tuned version of GPT-2, trained to classify tweets into sentiment categories. The model was fine-tuned on the **mteb/tweet_sentiment_extraction** dataset, which contains labeled tweets for sentiment analysis.

The model performs the task of classifying tweets into three sentiment categories:
- **Negative**: Label 0
- **Neutral**: Label 1
- **Positive**: Label 2

This model is suitable for analyzing sentiment in short-form text such as tweets, product reviews, or customer feedback.

## Intended Use

The model can be used for the following purposes:
- **Sentiment analysis** of short texts (e.g., tweets, reviews, feedback).
- **Customer feedback analysis** to classify sentiment in user comments.
- **Social media monitoring** to track the sentiment of public opinion about topics, brands, or products.

## How to Use

You can use the model with the Hugging Face `pipeline` API to classify the sentiment of a text input.

#### Example:

```python
from transformers import pipeline

# Load the fine-tuned model
classifier = pipeline("text-classification", model="riturajpandey739/gpt2-sentiment-analysis-tweets")

# Example text for sentiment classification
text = "This product is amazing! I absolutely love it."

# Get the sentiment prediction
result = classifier(text)

# Output the result
print(result)
# Example Output: [{'label': 'LABEL_2', 'score': 0.9976001381874084}]