NVIDIA-Merlin / NVIDIA-Merlin/Merlin

[FEA] Calculate minimum required GPU memory for dataloader

Open
#706 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

enhancement P1
Dominant language
Python
Stars
907
Forks
129
PR merge metrics
No merged PRs in 30d

Description

🚀 Feature request

Merlin dataloader loads training data in chunks into GPU memory. TensorFlow and Merlin dataloaders need to share the GPU memory. As a user, I want to calculate the minimum required GPU memory for the dataloader, that I know how to configure TensorFlow (providing it the maximum possible GPU memory). Currently, I run a lot of try and error to configure TensorFlow and run into OOM between TensorFlow and the data loader

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.

Research direction

Start by locating the Merlin dataloader's chunk-loading and GPU-memory configuration, then inspect how it shares memory with TensorFlow. The issue does not name files or tests; clarify the calculation inputs, supported configurations, and validation needed before implementation.

Written by the indexing model from the issue text.

Assessment

Tech stack
python, tensorflow
Domain
data, machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
25/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.