Logo
Porting TensorFlow 1 models to the IPU
Version: latest
  • 1. Introduction
  • 2. How to transfer an existing code base
    • 2.1. Preliminary analysis
    • 2.2. Planning the transfer
    • 2.3. Next steps
  • 3. ResNeXt inference example
    • 3.1. Abridged sample code
    • 3.2. Preliminaries: getting up and running
    • 3.3. Configuring the IPU
    • 3.4. Generating a report
    • 3.5. Infeeds and outfeeds
  • 4. Training with the estimator API
    • 4.1. Instantiate an estimator for the IPU
    • 4.2. Abridged code sample for the estimator
    • 4.3. Train and evaluate methods
  • 5. Scoping and determining unsupported operations
  • 6. Checking hardware availability
  • 7. ResNeXt full code example
  • 8. Trademarks & copyright
Porting TensorFlow 1 models to the IPU

Search help

Note: Searching from the top-level index page will search all documents. Searching from a specific document will search only that document.

  • Find an exact phrase: Wrap your search phrase in "" (double quotes) to only get results where the phrase is exactly matched. For example "PyTorch for the IPU" or "replicated tensor sharding"
  • Prefix query: Add an * (asterisk) at the end of any word to indicate a prefix query. This will return results containing all words with the specific prefix. For example tensor*
  • Fuzzy search: Use ~N (tilde followed by a number) at the end of any word for a fuzzy search. This will return results that are similar to the search word. N specifies the “edit distance” (fuzziness) of the match. For example Polibs~1
  • Words close to each other: ~N (tilde followed by a number) after a phrase (in quotes) returns results where the words are close to each other. N is the maximum number of positions allowed between matching words. For example "ipu version"~2
  • Logical operators. You can use the following logical operators in a search:
    • + signifies AND operation
    • | signifies OR operation
    • - negates a single word or phrase (returns results without that word or phrase)
    • () controls operator precedence

Porting TensorFlow 1 models to the IPU

This document is a practical guide to porting TensorFlow 1 models to the Poplar SDK for running on the IPU.

Contents

  • 1. Introduction
  • 2. How to transfer an existing code base
    • 2.1. Preliminary analysis
    • 2.2. Planning the transfer
    • 2.3. Next steps
  • 3. ResNeXt inference example
    • 3.1. Abridged sample code
    • 3.2. Preliminaries: getting up and running
    • 3.3. Configuring the IPU
    • 3.4. Generating a report
    • 3.5. Infeeds and outfeeds
  • 4. Training with the estimator API
    • 4.1. Instantiate an estimator for the IPU
    • 4.2. Abridged code sample for the estimator
    • 4.3. Train and evaluate methods
  • 5. Scoping and determining unsupported operations
  • 6. Checking hardware availability
  • 7. ResNeXt full code example
  • 8. Trademarks & copyright
Next

Revision 723a37cd.