README.md

metadata

title: StockPredictor
emoji: 📈
colorFrom: blue
colorTo: indigo
sdk: gradio
sdk_version: 5.29.0
app_file: app.py
pinned: false
license: mit

---

title: StockPredictor emoji: 📈 colorFrom: blue colorTo: indigo sdk: gradio sdk_version: "5.29.0" app_file: app.py pinned: false license: mit

📈 Stock Price Predictor

AI-powered stock price predictions using LSTM neural networks, technical analysis, and market sentiment analysis.

🚀 Features

⚡ Fast predictions

🧠 LSTM Deep Learning model

📊 Technical indicators (RSI, MACD, SMA, ROC)

💭 News sentiment analysis

💾 Prediction caching

📱 Gradio web interface

🌍 Supports US and Indian stocks

---

🛠️ Local Setup

1. Clone Repository

git clone https://github.com/your-username/stock-predictor.git cd stock-predictor

2. Create Virtual Environment

Windows

python -m venv venv venv\Scripts\activate

macOS/Linux

python3 -m venv venv source venv/bin/activate

---

3. Install Dependencies

pip install -r requirements.txt

---

4. Add Environment Variables

Create a .env file:

FINNHUB_API_KEY=your_api_key_here

Get a free API key from:

https://finnhub.io/register

---

5. Run Application

python app.py

The application will start at:

http://localhost:7860

---

🌐 Deploy on Hugging Face Spaces

Step 1: Create Space

1. Open:

https://huggingface.co/spaces

2. Click:

Create New Space

3. Configure:

SDK: Gradio

License: MIT

Visibility: Public or Private

---

Step 2: Upload Project Files

Upload the following files:

app.py model.py requirements.txt README.md .env.example .gitignore

---

Step 3: Add Repository Secrets

Go to:

Settings → Repository secrets

Add the following secret:

Name: FINNHUB_API_KEY Value: your_api_key

---

Step 4: Automatic Deployment

Hugging Face Spaces will automatically:

Install dependencies

Run app.py

Create a public deployment URL

---

📊 Supported Stock Symbols

US Stocks

AAPL MSFT GOOGL AMZN TSLA NVDA META NFLX

Indian Stocks

SBIN.NS INFY.NS RELIANCE.NS TCS.NS HDFCBANK.NS

---

🧠 Model Architecture

Input Features ↓ LSTM Layer (32 Units) ↓ Dense Layer ↓ Predicted Price

---

📈 Technical Indicators Used

RSI

MACD

SMA20

Volatility

ROC

Closing Price

---

⚡ Performance

Metric Value

Prediction Speed 1-2 seconds Cached Predictions 50-100ms Training Time ~2 minutes Memory Usage 2-3 GB Accuracy (MAPE) ~4.8% Directional Accuracy 76-77%

---

⚠️ Disclaimer

This project is for educational purposes only.

Not financial advice

Markets are unpredictable

Always conduct your own research

Invest responsibly

---

🔐 Security Notes

Never upload .env

Use HF Spaces Secrets for API keys

No user data stored

Uses public Yahoo Finance data

---

🐛 Common Issues

Configuration Error on HF Spaces

Use this exact YAML block at the top of README.md:

Important:

YAML must start at line 1

No spaces before ---

app_file is required

Use supported Gradio versions only

---

ModuleNotFoundError

pip install -r requirements.txt

---

No Data Found

Verify stock symbol format.

Example:

SBIN.NS

---

📁 Project Structure

stock-predictor/ │ ├── app.py ├── model.py ├── requirements.txt ├── README.md ├── .env.example └── .gitignore

---

📄 License

MIT License

---

❤️ Built With

Gradio

PyTorch

yfinance

Finnhub API

Python