Installation

TileFusion can be used as a lightweight C++ library with header-only usage, or it can be built as a Python library. You can choose to build either one.

Prerequisites

TileFusion requires:

C++20 host compiler
CUDA 12.0 or later
GCC version 10.0 or higher to support C++20 features

Download the repository:

git clone git@github.com:microsoft/TileFusion.git
cd TileFusion && git submodule update --init --recursive

Building the C++ Library

To build the project using the provided Makefile, simply run:

make

To run a single C++ unit test:

make unit_test_cpp CPP_UT=test_gemm

Building the Python Package

Build the wheel:
```
python setup.py build bdist_wheel
```
Clean the build:
```
python setup.py clean
```
Install the Python package in editable mode (recommended for development):
```
python setup.py develop
```
This allows you to edit the source code directly without needing to reinstall it repeatedly.

Running Unit Tests

Before running the Python unit tests, you need to build and install the Python package (see the Building the Python Package section).

Run a single Python unit test:
```
pytest tests/python/test_scatter_nd.py
```
Run all Python unit tests:
```
python setup.py pytests
```
Run all C++ unit tests:
```
python setup.py ctests
```