Workflow to test Triton with runtime dependencies installed with pip

intel / intel-xpu-backend-for-triton

OpenAI Triton backend for Intel® GPUs

MIT License

143 stars 44 forks source link

Workflow to test Triton with runtime dependencies installed with pip #2795

Open pbchekin opened 21 hours ago

pbchekin commented 21 hours ago

Instead of installing and initializing oneAPI components via standalone installer, the runtime dependencies for Triton can be installed with pip:

pip install intel-sycl-rt
pip install intel-pti
pip install mkl # optional if PyTorch compiled with dynamic MKL

export LD_LIBRARY_PATH=/path/to/the/installed/libs