Aidge Core

Aidge is a powerful open-source framework for optimizing and deploying deep neural networks on embedded systems. This is the core library required by every other module.

Quick Start

pip install aidge-core
python -c "import aidge_core ; print(f'Available backends:\n{aidge_core.get_keys_Conv2DOp()}')"

This should print:

Available backends:
{'export_serialize'}

[!NOTE] Pypi packages are experimental !

Build from Source

Prerequisite:

• Please review the global installation instructions before proceeding.

• Use the same virtual environment for all Aidge modules.

1. Python or C++ installation using setup scripts

[!TIP] Use Get-Help setup.ps1 (Windows) or ./setup.sh -h (Unix) for full documentation.

2. Python Installation using pip

Run these commands from the aidge_core/ directory:

# Standard install
pip install . -v

# Install with testing dependencies
pip install .[test] -v && pytest

Editable Install (Experimental)

For real-time development without re-installing:

pip install --no-build-isolation -ve . --config-settings=editable.rebuild=true -Cbuild-dir=build

3. C++ Installation using CMake

A CMakePresets.json is provided for standard configurations.

# Configure, Build, and Install
cmake --preset clang-debug
cmake --build --preset clang-debug
cmake --install

# Run C++ Tests
ctest --test-dir build/

[!TIP] Create a CMakeUserPresets.json to define your own local build configurations.

Static Build

For a static C++ build (e.g. for embedded integration without Python):

cmake -B build/static -DBUILD_SHARED_LIBS=OFF -DAIDGE_BUILD_PYTHON=OFF
cmake --build build/static

[!NOTE] Without pybind (AIDGE_BUILD_PYTHON=OFF), cross-module shared data (registrar, log levels, node IDs) relies on function-local static variables which require static linking. Use -DBUILD_SHARED_LIBS=OFF in this case. When building with pybind (AIDGE_BUILD_PYTHON=ON), shared libraries are the default and recommended.