6 comments (6 comments)0 reactions (0 reactions)0 assignees (0 assignees)Shell15,392 stars (15,392 stars)5,359 forks (5,359 forks)batch import
enhancementhelp wanted
Description
Pytorchnn only uses one gpu. It's very inefficient when training a large amount of text.
Contributor guide
- Tech stack
- pythonpytorch
- Domain
- machine learningbackend
- Issue type
- feature
- DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
- 5
- Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
- over 1 week
- 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.
- mostly clear
- Prerequisites
- Familiarity with PyTorchUnderstanding of multi GPU trainingExperience with Kaldi's codebaseKnowledge of CUDA
- Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
- 15
- Research direction
- The issue requests extending Kaldi's Pytorchnn module to utilize multiple GPUs. Start by examining the current single GPU implementation in the source code (likely under src/pytorchnn/). Research PyTorch's DataParallel and DistributedDataParallel modules. Check the issue comments for any suggested approaches or pointers from maintainers. Then propose a design that integrates multi GPU support without breaking existing functionality.