1. Installation

This guide will walk you through installing TurboGuard and all its dependencies on your system.

1.1. System Requirements

Before installing TurboGuard, ensure your system meets these requirements:

Operating System - Windows - macOS - Linux

Python - Python 3.8 or higher - pip package manager

Hardware - Minimum 6GB RAM (8GB recommended) - GPU support optional but recommended for training

1.2. Installation Methods

1.2.2. Method 2: Development Installation

For contributors or advanced users who want to modify the code:

git clone https://github.com/mouradboutrid/TurboGuard.git
cd TurboGuard
pip install -e .

This installs TurboGuard in β€œeditable” mode, so changes to the source code are immediately reflected.

1.3. Required Dependencies

TurboGuard requires the following main packages:

Core Dependencies

tensorflow>=2.8.0
numpy>=1.21.0
pandas>=1.3.0
scikit-learn>=1.0.0
matplotlib>=3.5.0
seaborn>=0.11.0

Dashboard Dependencies

streamlit>=1.12.0
plotly>=5.10.0
altair>=4.2.0

Utility Dependencies

tqdm>=4.64.0
pyyaml>=6.0
joblib>=1.1.0

1.4. GPU Support (Optional)

To enable GPU acceleration for model training:

NVIDIA GPU Setup

  1. Install CUDA Toolkit (11.2 or later):

    Download from NVIDIA CUDA Toolkit

  2. Install cuDNN:

    Download from NVIDIA cuDNN

  3. Install TensorFlow GPU support:

    pip install tensorflow[and-cuda]

Verify GPU Installation

import tensorflow as tf
print("GPU Available: ", tf.config.list_physical_devices('GPU'))

1.5. Verify Installation

Test your installation with these verification steps:

1. Import Test

# Test core imports
import tensorflow as tf
import pandas as pd
import numpy as np
import streamlit as st

print("βœ… All core dependencies imported successfully!")

2. TurboGuard Import Test

from src.LSTM_AutoEncoder.data_loader import CMAPSSDataLoader
from src.LSTM_AutoEncoder.lstm_autoencoder import LSTMAutoencoder
from src.Forecasting_LSTM.forecasting_lstm import ForecastingLSTM

print("βœ… TurboGuard modules imported successfully!")

3. Dashboard Test

streamlit run app/app.py

If successful, you should see:

You can now view your Streamlit app in your browser.
Local URL: http://localhost:8501

1.6. Common Installation Issues

Issue 1: TensorFlow Installation Fails

Error: ERROR: Could not find a version that satisfies the requirement tensorflow

Solution: - Ensure Python version is 3.8-3.11 - Update pip: pip install --upgrade pip - Try: pip install tensorflow --upgrade

Issue 2: CUDA/GPU Issues

Error: Could not load dynamic library 'libcudart.so.11.0'

Solution: - Verify CUDA installation - Check CUDA version compatibility with TensorFlow - Install matching cuDNN version

Issue 3: Memory Issues During Installation

Error: MemoryError during package installation

Solution: - Close other applications - Install packages one by one - Use: pip install --no-cache-dir -r requirements.txt

Issue 4: Streamlit Port Already in Use

Error: OSError: [Errno 48] Address already in use

Solution: - Use different port: streamlit run app/app.py --server.port 8502 - Kill existing process on port 8501

1.7. Next Steps

Once installation is complete:

  1. βœ… Continue to Quick Start to launch your first TurboGuard session

  2. πŸ“Š Explore the interactive dashboard

  3. πŸ€– Build your first model in Your First Model

1.8. Need Help?

If you encounter issues not covered here:

  • πŸ› Report bugs: GitHub Issues

  • πŸ’¬ Ask questions: Create a discussion on GitHub

  • πŸ“– Check docs: Refer to our detailed API documentation

Congratulations! You’re ready to start using TurboGuard! πŸŽ‰