From e5b7cff5fe65eac9e54ba88fa3935b3270db0207 Mon Sep 17 00:00:00 2001 From: "Wang, Yi" Date: Thu, 22 Sep 2022 21:20:15 +0800 Subject: [PATCH] update perf_train_cpu_many doc (#19151) Signed-off-by: Wang, Yi A Signed-off-by: Wang, Yi A --- docs/source/en/perf_train_cpu_many.mdx | 20 +++++++++++++++++--- 1 file changed, 17 insertions(+), 3 deletions(-) diff --git a/docs/source/en/perf_train_cpu_many.mdx b/docs/source/en/perf_train_cpu_many.mdx index f4f7796574..9e7bd61467 100644 --- a/docs/source/en/perf_train_cpu_many.mdx +++ b/docs/source/en/perf_train_cpu_many.mdx @@ -38,16 +38,30 @@ where `{pytorch_version}` should be your PyTorch version, for instance 1.12.0. Check more approaches for [oneccl_bind_pt installation](https://github.com/intel/torch-ccl). Versions of oneCCL and PyTorch must match. + + +oneccl_bindings_for_pytorch 1.12.0 prebuilt wheel does not work with PyTorch 1.12.1 (it is for PyTorch 1.12.0) + + + ## Intel® MPI library Use this standards-based MPI implementation to deliver flexible, efficient, scalable cluster messaging on Intel® architecture. This component is part of the Intel® oneAPI HPC Toolkit. -It can be installed via [MPI](https://www.intel.com/content/www/us/en/developer/articles/tool/oneapi-standalone-components.html#mpi). -Please set the environment by following command before using it. +oneccl_bindings_for_pytorch is installed along with the MPI tool set. Need to source the environment before using it. +for Intel® oneCCL 1.12.0 ``` -source /opt/intel/oneapi/setvars.sh +oneccl_bindings_for_pytorch_path=$(python -c "from oneccl_bindings_for_pytorch import cwd; print(cwd)") +source $oneccl_bindings_for_pytorch_path/env/setvars.sh ``` +for Intel® oneCCL whose version < 1.12.0 +``` +torch_ccl_path=$(python -c "import torch; import torch_ccl; import os; print(os.path.abspath(os.path.dirname(torch_ccl.__file__)))") +source $torch_ccl_path/env/setvars.sh +``` + + The following "Usage in Trainer" takes mpirun in Intel® MPI library as an example.