pytorch/ignite

Debug mode for ignite

Open

#1,992 opened on May 7, 2021

View on GitHub
 (15 comments) (2 reactions) (0 assignees)Python (4,313 stars) (602 forks)batch import
enhancementhelp wanted

Description

🚀 Feature

The idea is to provide a debug mode as discussed here: https://github.com/pytorch/ignite/issues/1989#issuecomment-833643158.

All the possible options can be discussed here. We could add an helper to track the workflow of handlers from the engine run. On the other hand, some specific debug tools are needed to track the values of PyTorch objects (LR, loss, grads, etc.). Note that the preferred way to extend ignite is handlers' design so let's create a new set of tools for the debugging.

Contributor guide

Tech stack
python
Domain
machine learningdeveloper experience
Issue type
feature
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
3
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-2 days
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
PythonPyTorchIgniteDebugging
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
50
Research direction
Review the referenced discussion in issue #1989 for context on the desired debug features. Analyze the Ignite handler pattern to design a debug handler that logs PyTorch objects like loss and gradients. Consider creating a separate module for debugging tools that can be attached to the engine. The implementation should follow the existing handler design to maintain consistency.