The Intel OneAPI and AI Analytics Toolkit (AI Kit) provides familiar Python tools and frameworks to accelerate end-to-end data science and analytics pipelines on Intel® architecture. The components are built using oneAPI libraries for low-level compute optimizations. This toolkit maximizes performance from preprocessing through machine learning and provides interoperability for efficient model development.
Many pre-built AI Kit Docker images run natively in Domino. See https://hub.docker.com/r/intel/oneapi-aikit for a list of available images on Dockerhub.
-
Create a new environment from a custom image.
-
Paste or type the Docker registry path of the image that you want to use in the FROM field. For example,
intel/oneapi-aikit:latest.The environment includes the following packages:
intel-aikit-getting-started intel-basekit-getting-started intel-level-zero-gpu intel-oneapi-advisor intel-oneapi-ccl-devel intel-oneapi-common-licensing intel-oneapi-common-vars intel-oneapi-compiler-dpcpp-cpp intel-oneapi-dal-devel intel-oneapi-dev-utilities intel-oneapi-dnnl-devel intel-oneapi-dpcpp-debugger intel-oneapi-ippcp-devel intel-oneapi-ipp-devel intel-oneapi-libdpstd-devel intel-oneapi-mkl-devel intel-oneapi-model-zoo intel-oneapi-onevpl-devel intel-oneapi-python intel-oneapi-pytorch intel-oneapi-tbb-devel intel-oneapi-tensorflow intel-oneapi-vtune intel-opencl-icd level-zero level-zero-dev
-
After the environment is built, create a new workspace.
See the following for more information: