tqchen / tlcpack

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

tlcpack

Conda-CPU-Nightly Conda-GPU-Nightly Wheel-WinMac-Nightly Wheel-ManyLinux-Nightly Prune-Nightly

Tensor learning compiler binary distribution package.

Github Actions

We use github action to build wheel and conda packages nightly.

Checkout .github/workflows

Build Process

  1. Build docker images
./docker/build_image.sh <CONTAINER_TYPE>

CONTAINER_NAME: Type of the docker container used to build wheels, e.g., (cpu|cpu_aarch64|cu100|cu101|cu102)
  1. Checkout tvm and sync version
git clone https://github.com/apache/tvm --recursive
# synchronize the package version
python common/sync_package.py [tlcpack|tlcpack-nightly]

The nightly will point to the latest main, tlcpack will point to a stable build hashtag defined in common/sync_package.py

  1. Build tlcpack manylinux wheels.
./docker/bash.sh [docker-image] ./wheel/build_wheel_manylinux.sh --cuda none

To build wheels for a specific CUDA version, for example, CUDA 11.1, run

./docker/bash.sh [docker-image] ./wheel/build_wheel_manylinux.sh --cuda 11.1

The docker image is built in step 1 and needs to match the cuda version.

  1. Get the wheels

The wheels are now available in

./tvm/python/repaired_wheels

About

License:Apache License 2.0


Languages

Language:Groovy 36.3%Language:Python 32.4%Language:Shell 28.6%Language:Batchfile 1.7%Language:CMake 1.0%