Skip to content
This repository has been archived by the owner on Apr 6, 2023. It is now read-only.

Latest commit

 

History

History
 
 

popdist

Folders and files

NameName
Last commit message
Last commit date

parent directory

..
test
test
 
 
README.md
README.md
 
 
popdist_training.py
popdist_training.py
 
 

README.md

Graphcore

PopDist example

File structure

  • popdist_training.py Example training script with PopDist support.
  • test Integration tests for this example.
  • README.md This file.

How to use this demo

  1. Prepare the TensorFlow environment.

    Install the Poplar SDK following the instructions in the Getting Started guide for your IPU system. Make sure to source the enable.sh script for Poplar and activate a Python virtualenv with either a TensorFlow 1 or TensorFlow 2 wheel from the Poplar SDK installed (use the version appropriate to your operating system and processor).

  2. Run the script using PopRun. The number of instances and replicas are provided as command-line arguments. Example:

    poprun --num-instances=2 --num-replicas=4 python3 popdist_training.py