Shoriful025/crypto_trend_predictor

Crypto Trend Predictor

Overview

This BERT-based model predicts cryptocurrency market trends (bearish, bullish, or neutral) based on news articles, tweets, or market summaries. It was fine-tuned on a dataset of historical crypto news and price movements.

Model Architecture

• Base Model: BERT-base-uncased
• Layers: 12
• Hidden Size: 768
• Attention Heads: 12
• Fine-tuned for multi-class classification (bearish/bullish/neutral)

Intended Use

Ideal for analyzing crypto-related text to forecast short-term market trends, assisting traders or analysts in decision-making.

Limitations

• Predictions are based on text sentiment and may not account for external factors like regulations or economic events.
• Performance degrades on non-English text or highly technical jargon not seen in training.
• Not financial advice; use at your own risk.

Example Code

from transformers import pipeline

predictor = pipeline("text-classification", model="user/crypto-trend-predictor")
result = predictor("Bitcoin surges after ETF approval.")
print(result)
# [{'label': 'BULLISH', 'score': 0.95}]