pytorch/ignite

Add Expected Calibration Error metrics and derived

Open

#1,009 opened on May 4, 2020

View on GitHub
 (10 comments) (0 reactions) (0 assignees)Python (602 forks)batch import
enhancementhelp wanted

Repository metrics

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

Description

🚀 Feature

Idea is to provide Expected Calibration Error metrics as coded in Baal : https://github.com/ElementAI/baal/blob/master/src/baal/utils/metrics.py#L88

cc @Dref360

Contributor guide

Research direction
Implement Expected Calibration Error (ECE) and related metrics as described in the Baal library (https://github.com/ElementAI/baal/blob/master/src/baal/utils/metrics.py#L88). The implementation should follow PyTorch Ignite's metrics API pattern. Consider adapting the existing code structure in ignite/metrics. The main task is to port the ECE computation, which involves binning confidence scores and computing accuracy per bin. Check for any existing calibration related utilities in the repository.
Tech stack
python
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.
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.
Clear
Prerequisites
PyTorch basicsCalibration metrics concepts
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
65