pytorch/ignite

Enhancements to EarlyStopping handler

Open

#3,411 opened on Jun 4, 2025

View on GitHub
 (19 comments) (0 reactions) (0 assignees)Python (602 forks)batch import
enhancementhelp wantedmodule: handlers

Repository metrics

Stars
 (4,313 stars)
PR merge metrics
 (Avg merge 15d 11h) (17 merged PRs in 30d)

Description

🚀 Feature

I would propose several enhancements to the EarlyStopping handler.

  1. Introduce a new parameter, min_delta_mode, mimicking the parameter threshold_mode of ReduceLROnPlateau. If min_delta_mode="abs" (default) everything works as of now, while if min_delta_mode="rel" the score comparison becomes score >= best_score (1 + min_delta).
  2. Introduce a new parameter mode that work's as in ReduceLROnPlateau, allowing to handle validation metrics to be minimized.

Other ideas I'd like to discuss are:

  • Natively implementing min_epochs/min_evals, preventing EarlyStopping from terminating until it has been evaluated at least X times. I know this could be addressed as in #3317 but this solution might be more compact?
  • It would be nice to have the same threshold and threshold_mode nomenclature as ReduceLROnPlateau, but I realize this would probably break older code.
  • I personally find the parameter name cumulative_delta quite counter-intuitive. When I first read about that parameter, I understood it was taking the cumulative sum of the improvements and comparing this cumulative sum to min_delta.

Contributor guide

Research direction
Review the existing EarlyStopping handler in ignite/handlers/early stopping.py. Study the ReduceLROnPlateau implementation in PyTorch for mode and threshold mode parameters. Consider the discussion in this issue and linked issue #3317. Propose a plan for implementing the suggested parameters while maintaining backward compatibility.
Tech stack
pythonpytorch
Domain
machine learning
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.
3-5 days
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Mostly clear
Prerequisites
PyTorch Ignite basicsReduceLROnPlateau API
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
50