₿ Bitcoin Price Prediction using LSTM (Deep Learning)

This project predicts future Bitcoin (BTC) prices using LSTM (Long Short-Term Memory) — a deep learning model specialized in time series forecasting. The model learns patterns in historical BTC data to make future price predictions.

---

📁 Dataset

The dataset includes historical Bitcoin prices with key columns:

• Date
• Open
• High
• Low
• Close
• Volume

The Close price is used as the target variable for forecasting.

---

📝 Topics Covered

• Time Series Forecasting
• Deep Learning with LSTM Networks
• Data Normalization using MinMaxScaler
• Sequence Generation for Time Series Input
• Building and Training LSTM Models
• Visualizing Predictions vs Actuals
• Model Evaluation using Mean Absolute Error (MAE)

---

⚙️ Technologies Used

• Python
• Pandas, NumPy
• Matplotlib, Seaborn
• TensorFlow / Keras
• Scikit-learn
• Jupyter Notebook

---

🔍 Project Highlights

• Built an LSTM neural network for time series forecasting.
• Converted historical BTC prices into time-based sequences.
• Normalized and reshaped data for LSTM input compatibility.
• Generated future price predictions and compared with actuals.
• Evaluated prediction performance using Mean Absolute Error (MAE).

---

🎓 What I Learned

• Preparing financial time series data for deep learning models.
• Sequence windowing and data shaping for LSTM training.
• Importance of normalization in neural network performance.
• Interpreting MAE as a performance metric in forecasting tasks.

---