Skip to content

Build pytorch from sources

Automatically generated README for this automation recipe: install-pytorch-from-src

Category: Compiler automation

License: Apache 2.0

  • CM meta description for this script: _cm.json
  • Output cached? True

Reuse this script in your project

Install MLCommons CM automation meta-framework

Pull CM repository with this automation recipe (CM script)

cm pull repo mlcommons@cm4mlops

cmr "install get src from.src pytorch src-pytorch" --help

Run this script

Run this script via CLI
cm run script --tags=install,get,src,from.src,pytorch,src-pytorch[,variations] 
Run this script via CLI (alternative)
cmr "install get src from.src pytorch src-pytorch [variations]" 
Run this script from Python
import cmind

r = cmind.access({'action':'run'
              'automation':'script',
              'tags':'install,get,src,from.src,pytorch,src-pytorch'
              'out':'con',
              ...
              (other input keys for this script)
              ...
             })

if r['return']>0:
    print (r['error'])
Run this script via Docker (beta)
cm docker script "install get src from.src pytorch src-pytorch[variations]" 

Variations

  • No group (any combination of variations can be selected)

    Click here to expand this section.

    • _branch.#
      • ENV variables:
        • CM_GIT_CHECKOUT: #
    • _cuda
      • ENV variables:
        • CUDA_HOME: <<<CM_CUDA_INSTALLED_PATH>>>
        • CUDNN_LIBRARY_PATH: <<<CM_CUDA_PATH_LIB_CUDNN>>>
        • CUDNN_INCLUDE_PATH: <<<CM_CUDA_PATH_INCLUDE_CUDNN>>>
        • CUDA_NVCC_EXECUTABLE: <<<CM_NVCC_BIN_WITH_PATH>>>
        • USE_CUDA: 1
        • USE_CUDNN: 1
        • TORCH_CUDA_ARCH_LIST: Ampere Ada Hopper
        • TORCH_CXX_FLAGS: -D_GLIBCXX_USE_CXX11_ABI=1
    • _for-intel-mlperf-inference-v3.1-bert
      • ENV variables:
        • CM_CONDA_ENV: yes
        • CM_MLPERF_INFERENCE_INTEL: yes
        • USE_CUDA: 0
    • _for-nvidia-mlperf-inference-v3.1
    • _for-nvidia-mlperf-inference-v4.0
    • _sha.#
      • ENV variables:
        • CM_GIT_CHECKOUT_SHA: #
    • _tag.#
      • ENV variables:
        • CM_GIT_CHECKOUT_TAG: #
  • Group "repo"

    Click here to expand this section.

    • _repo.#
      • ENV variables:
        • CM_GIT_URL: #
    • _repo.https://github.com/pytorch/pytorch (default)
      • ENV variables:
        • CM_GIT_URL: https://github.com/pytorch/pytorch
Default variations

_repo.https://github.com/pytorch/pytorch

Native script being run

No run file exists for Windows


Script output

cmr "install get src from.src pytorch src-pytorch [variations]"  -j