AIKA Installation Guide START HERE

New to AIKA? Follow this guide to set up your development environment. AIKA is a C++ neural network framework with Python bindings.

⏱️ Estimated time: 10-15 minutes | 📊 Skill level: Beginner-friendly

Prerequisites

Python 3.8+ with pip
CMake 3.30+
C++ compiler with C++20 support (GCC, Clang, or MSVC)
pybind11 - Installed automatically via pip, or manually:
- macOS: brew install pybind11
- Linux: apt-get install pybind11-dev or pip install pybind11
- Windows: pip install pybind11

Installation Steps

# Step 1: Clone the AIKA repository
git clone https://github.com/aika-algorithm/aika.git
cd aika-cpp

# Step 2: Create and activate a virtual environment
python3 -m venv .venv
source .venv/bin/activate  # On Windows: .venv\Scripts\activate

# Step 3: Install Python package in development mode
pip install -e .

# Step 4: Build the C++ core (out-of-source build)
mkdir -p build && cd build
cmake ..
cmake --build . --target install

# Step 5: Verify installation
cd ..
python tests/python/fields/subtraction-test.py

If the test runs successfully and prints computation results, AIKA is installed correctly.

Project Structure

After installation, your project will have the following structure:

src/ - C++ source code
include/ - C++ header files
python/ - Python helper modules (networks, types, utils, examples)
tests/ - Test suite (cpp/ and python/ subdirectories)
build/ - Build artifacts (created during compilation)
specs/ - Formal specifications and documentation

Running Tests

To verify your installation, run the test suite:

# Python tests (Fields Module)
python -m unittest discover tests/python/fields

# Python tests (Network Module)
python -m unittest discover tests/python

# C++ tests
cd build && ctest

Troubleshooting

If you encounter any issues during installation, check the following:

Ensure all prerequisites are installed and accessible from the command line.
Verify CMake is detecting your C++20-compatible compiler correctly.
Check that Python and pip are from your virtual environment (which python).
If pybind11 is not found, install it manually: pip install pybind11
On macOS, you may need to install Xcode Command Line Tools: xcode-select --install
For additional help, visit the GitHub Issues Page.

Next Steps

✅ Installation Complete! You're ready to start building with AIKA.

Continue to Tutorial →

Learn how to build your first AIKA neural network with our step-by-step Tutorial, or explore Why AIKA? to understand what makes it different.