🔮 Predict Stock Prices Using Deep Learning
Example symbols:
BTC-USD
RELIANCE
TCS
INFY
HDFCBANK
Training CNN+LSTM model... This may take a minute.
Fetching 4 years of historical data and training deep learning model
Model Accuracy
—
MAE
—
RMSE
—
R² Score
—
Day 1
—
Day 2
—
Day 3
—
Day 4
—
Day 5
—
Price Prediction Chart
5-Day Price Forecast
| Date | Predicted Price |
|---|
About CNN + LSTM Model
This deep learning model combines Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks to capture both local patterns and long-term dependencies in stock price data. The CNN layer extracts local features from the input sequences, while the LSTM layer learns temporal patterns over longer periods. The model is trained on 4 years of historical data using a 60-day lookback window to predict the next 5 days of closing prices.
Model Architecture
- • Conv1D (64 filters, kernel_size=3)
- • MaxPooling1D (pool_size=2)
- • Dropout (0.2)
- • LSTM (100 units)
- • Dense (50 units)
- • Output (1 unit)
Training Configuration
- • Epochs: 20
- • Batch Size: 32
- • Lookback: 60 days
- • Optimizer: Adam
- • Loss: Mean Squared Error