financial-sentiment-distilbert

financial-sentiment-distilbert

Model Description

DistilBERT-based financial sentiment analysis model trained on balanced dataset

This model is fine-tuned from distilbert-base-uncased for financial sentiment analysis, capable of classifying financial text into three categories:

• Bearish (0): Negative financial sentiment
• Neutral (1): Neutral financial sentiment
• Bullish (2): Positive financial sentiment

Model Performance

• Accuracy: 0.797
• Dataset: Twitter Financial News Sentiment
• Base Model: distilbert-base-uncased

Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch

# Load model and tokenizer
tokenizer = AutoTokenizer.from_pretrained("codealchemist01/financial-sentiment-distilbert")
model = AutoModelForSequenceClassification.from_pretrained("codealchemist01/financial-sentiment-distilbert")

# Example usage
text = "Apple stock is showing strong growth potential"
inputs = tokenizer(text, return_tensors="pt", truncation=True, padding=True)

with torch.no_grad():
    outputs = model(**inputs)
    predictions = torch.nn.functional.softmax(outputs.logits, dim=-1)
    predicted_class = torch.argmax(predictions, dim=-1).item()

# Labels: 0=Bearish, 1=Neutral, 2=Bullish
labels = ["Bearish", "Neutral", "Bullish"]
print(f"Prediction: {labels[predicted_class]}")

Training Details

• Training Dataset: Twitter Financial News Sentiment
• Training Framework: Transformers
• Optimization: AdamW
• Hardware: RTX GPU

Limitations

This model is specifically trained for financial sentiment analysis and may not perform well on general sentiment analysis tasks.

Citation

If you use this model, please cite:

@misc{financial-sentiment-distilbert,
  author = {CodeAlchemist01},
  title = {financial-sentiment-distilbert},
  year = {2024},
  publisher = {Hugging Face},
  url = {https://huggingface.co/codealchemist01/financial-sentiment-distilbert}
}