Skip to content

EFS-OpenSource/calibration-framework

View on GitHub ↗Related repositories →

The net:cal calibration framework is a Python 3 library for measuring and mitigating miscalibration of uncertainty estimates, e.g., by a neural network.

active 2023-08-182026-04-16 (UTC)

Partial coverage20,441 / 26,198 hourly files (78%) · 2 absent upstream · 5,753 failed, retryable2023-08-172026-08-13 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
91
Pushes
2
Pull requests
0
Issues
12
Stars
67
Forks
4

Activity over time

Daily event counts in the loaded window

Line chart, 973 days from 2023-08-18 to 2026-04-16. Pushes: 2 total, peak 1 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 12 total, peak 1 in a day. Comments: 6 total, peak 2 in a day. Stars: 67 total, peak 2 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

ContributorContributionsPushesPRsComments
fabiankueppers8203
Ruhallah935003
Niko-k981000
Beemd1000
soberguo1000
venk-na1000
anishkanabar1000
YuanyuanLi961000
vitawasalreadytaken1000

Recent activity

Latest issues, pull requests and releases

  • Issue comment#63fabiankueppers2025-07-09 12:28
    Clarification needed: Use of network logits vs. predicted probabilities for calibration input in the framework
  • Issue#63venk-na2025-07-03 09:45
    Clarification needed: Use of network logits vs. predicted probabilities for calibration input in the framework
  • Issue#62vitawasalreadytaken2025-06-09 14:03
    Dependence on tikzplotlib makes netcal.presentation unusable
  • Issue#60soberguo2024-08-18 08:34
    Why do I already have torch installed, but when I install this package with pip, it still forces the torch to be updated?
  • Issue#58Beemd2024-07-20 01:45
    Issue with Probabilistic Regression GPBeta
  • Issue comment#57Ruhallah932024-05-26 09:59
    ENIR
  • Issue#56Ruhallah932024-05-12 10:07
    NaN outputs
  • Issue comment#55Ruhallah932024-05-04 19:21
    TemperatureScaling
  • Issue comment#55Ruhallah932024-05-04 18:59
    TemperatureScaling
  • Issue comment#55fabiankueppers2024-05-02 06:28
    TemperatureScaling
  • Issue#55fabiankueppers2024-05-02 06:28
    TemperatureScaling
  • Issue#55Ruhallah932024-04-20 14:18
    TemperatureScaling
  • Issue#51fabiankueppers2023-12-18 07:40
    Question about input range of multivariate confidence calibration
  • Issue#50anishkanabar2023-11-21 23:52
    SystemError: PY_SSIZE_T_CLEAN macro must be defined for '#' formats
  • Issue#48YuanyuanLi962023-10-26 23:14
    TemperatureScaling().transform() return input confidences.
  • Issue#47Niko-k982023-10-11 20:03
    inference with calibrated model
  • Issue comment#46fabiankueppers2023-08-30 07:28
    How to evaluate using D-ECE
  • Issue#46fabiankueppers2023-08-30 07:28
    How to evaluate using D-ECE

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 67 stars here means stars gained during the window, not the repo's star count.