tensorflow / tensorflow/recommenders

Does tensorflow recommenders support terabyte-scale embedding table?

Open
#127 1 comment 0 reactions 1 assignee View on GitHub

@maciejkula is already working on this.

Since Oct 26, 2020.

question
Dominant language
Python
Stars
2k
Forks
300
No merged PRs in 30d

Description

Dear recommenders authors,

Suppose the number of features is 10^9, and the embedding size is 1000, then the size of the embedding table is 10^12, which is terabyte scale. GPU or CPU, does recommenders support an embedding table this large? And, would parameter server strategy be the only strategy that I should use?

Thank you for your time!

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Assessment

This issue has not been assessed yet.

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.