rstudio/tensorflow

Best practices for using intel optimizations for tensorflow

Open

#453 opened on Feb 7, 2021

View on GitHub
 (2 comments) (0 reactions) (0 assignees)R (1,341 stars) (318 forks)batch import
docshelp wanted

Description

It would be great if we could get some documentation for using the intel optimizations for tensorflow in R.

The first step in getting this working is to get reticulate to point to a conda environment that has the intel ai toolkit (and corresponding versions of tensorflow installed), but I believe that's not sufficient. How might we change, from R, the settings described (here, here, here, and here)?

Contributor guide

Tech stack
rtensorflowpython
Domain
documentation
Issue type
documentation
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-3 hours
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
stale
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
needs investigation
Prerequisites
R programmingTensorFlow basicsconda environmentsIntel AI toolkit
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
25
Research direction
First, review the linked Intel AI tutorials (e.g., IntelAI/models docs) to understand how to set environment variables and thread settings for TensorFlow on Intel CPUs. Then, investigate how reticulate can point to a conda environment with the Intel AI toolkit and test whether environment variables set in R (via Sys.setenv) affect TensorFlow behavior. Finally, document the steps for R users, including recommended conda environment setup and any performance hints.