numba-dppy extends Numba with a new backend to support compilation
for Intel CPU and GPU architectures.
For more information about Numba, see the Numba homepage: http://numba.pydata.org.
Note: numba-dppy requires patched version of Numba.
See https://github.com/IntelPython/numba.
For more information about dpCtl, see the dpCtl homepage: https://intelpython.github.io/dpctl/
For more information about dpNP, see the dpNP homepage: https://intelpython.github.io/dpnp/
- numba 0.52.* (IntelPython/numba)
- dpCtl 0.6.*
- dpNP >=0.5.1 (optional)
- llvm-spirv (SPIRV generation from LLVM IR)
- llvmdev (LLVM IR generation)
- spirv-tools
- scipy (for testing)
dpPy is a proof-of-concept backend for NUMBA to support compilation for Intel CPU and GPU architectures. The present implementation of dpPy is based on OpenCL 2.1, but is likely to change in the future to rely on Sycl/DPC++ or Intel Level-0 driver API.
Use setup.py or conda (see conda-recipe).
See folder numba_dppy/tests.
Run tests:
python -m pytest --pyargs numba_dppy.testsor
pytestSee folder numba_dppy/examples.
Run examples:
python numba_dppy/examples/sum.pyPlease follow instructions in the debugging.md
Please use https://github.com/IntelPython/numba-dppy/issues to report issues and bugs.
Read this guide for additional features INDEX.md