pytorch / pytorch/TensorRT

📖 [Story] Caching and Compile time improvements

Open
#2,684 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Story Story: Runtime & Memory & Serialization
Dominant language
Python
Stars
3k
Forks
410
Avg merge
3d 18h
Merged PRs (30d)
78

Description

TL;DR

There are a collection of features which work together to improve compile times

Goal(s)

We want to reduce the time it takes to compile models. Some parts are out of our control (dynamo graph capture) but we can improve on the TensorRT side.

Tasks

### Tasks
- [ ] https://github.com/pytorch/TensorRT/issues/2674
- [ ] Weight Refit
- [ ] Engine Caching

Additional context

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 reviewing linked issue 2674 and the TensorRT repository context for the two unlinked tasks, Weight Refit and Engine Caching. The story is complete when the compile-time improvements are implemented and their performance impact is demonstrated, but it does not identify files, entry points, or tests.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
compilers, machine-learning
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
15/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.