Data

Tools

Indexes

Applications

Experiments

Open Source Science

diametrics_lite

DiaMetrics (lite) is a web-based educational resource for exploring important concepts regarding binary classification (and its evaluation). While the "full" version is primarily intended for independent learning, the lite version is intended for use with direct instruction (e.g. in the context of lectures or talks).

https://github.com/lrosenplaenter/diametrics_lite

Science Score: 67.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • ✓
    CITATION.cff file
    Found CITATION.cff file
  • ✓
    codemeta.json file
    Found codemeta.json file
  • ✓
    .zenodo.json file
    Found .zenodo.json file
  • ✓
    DOI references
    Found 2 DOI reference(s) in README
  • ✓
    Academic publication links
    Links to: zenodo.org
  • ○
    Academic email domains
  • ○
    Institutional organization owner
  • ○
    JOSS paper metadata
  • ○
    Scientific vocabulary similarity
    Low similarity (11.7%) to scientific vocabulary
Last synced: 9 months ago · JSON representation ·

Repository

DiaMetrics (lite) is a web-based educational resource for exploring important concepts regarding binary classification (and its evaluation). While the "full" version is primarily intended for independent learning, the lite version is intended for use with direct instruction (e.g. in the context of lectures or talks).

Basic Info
Statistics
  • Stars: 0
  • Watchers: 1
  • Forks: 0
  • Open Issues: 0
  • Releases: 3
Created over 2 years ago · Last pushed about 1 year ago
Metadata Files
Readme License Citation

README.md

DiaMetrics (lite)

DiaMetrics (lite) is a web-based educational resource for exploring important concepts regarding binary classification (and its evaluation). While the "full" version is primarily intended for independent learning, the lite version is intended for use with direct instruction (e.g. in the context of lectures or talks).

DOI

Installation & Usage

No installation is necessary. Open DiaMetrics (lite) directly. Alternatively: Simply download and open index.html in your browser or store all files on your web server.

If you actually publish DiaMetrics elsewhere (e.g. on your website, in another project, etc.) I would be happy to be notified!

One and two-dimensional separators

New in version v1.1.0: In addition to the two-dimensional separator, a one-dimensional separator ("cut-off") can now also be used. You can switch between the separator modes in the settings or by setting parameters in the url.

"Fairness-Mode"

In "fairness mode", the portions of groups A and B are additionally accentuated to illustrate aspects of fairness for subgroups in binary classification. The fairness mode can be activated via the settings or parameters in the url (see below).

Customisation of the labelling of scales and groups, switching on/off "Fairness-Mode" & Separator-Mode

The naming of the groups and axes (or scales) can be changed via the settings directly on the page, or via parameters in the url. When the labelling is changed, a new url is automatically generated that contains the updated parameters and can be copied directly from the settings and can be passed on to third parties.

The parameters are: * vara, varb, varc for the labels of groups A, B and C (Group C gets highlighted in fairness-mode). * mx, my for the labels for the scales/ axes. * fair can be set to "true" or "false", to switch on or off fairness mode. * sepm can be set to "1d" for a one-dimensional separator ("cut-off") or to "2d" two-dimensional separator ("classification")

Example: https://lrosenplaenter.github.io/DiaMetrics_lite/?vara=Test&fair=true switches on the fairness mode and sets the label for group A to "Test".

Customisation of the data

The data displayed in the plot is defined via data.js. The data is organised in an object ext_data that contains three arrays (data_a; data_b; data_overlay). Each of the arrays contains 200 objects with x-y data pairs, which can be customised, added or removed as required.

data_a contains the data pairs for group A which are displayed in teal, and data_b contains the data pairs for group B which are displayed in grapefruit.

Note: data_overlay contains the data pairs that are to be highlighted in fairness mode. These should match individual data points from data_a and data_b accordingly.

Versions

There are two other versions of DiaMetrics:

  • DiaMetrics (web | repository) is the full version of DiaMetrics which includes examples and more detailed explanations.
  • DiaMetrics_DE (web | repository) is the German version of DiaMetrics.

Contributing

If you have found any bug or typo (or anything else that doesn't seem right) in DiaMetrics (lite), I would be happy to be notified directly. You can also open an issue on github.

Note: Issues and pull requests are not actively monitored on a regular basis.

Authors, citation and acknowledgments

DiaMetrics was developed by Leon Rosenplänter for teaching at the Department of Psychological Diagnostics (Prof. Dr. Martin Kersting), Justus-Liebig-University Giessen, Germany. Many thanks to S. Bender and D. Bonarius for their essential feedback and corrections, greatly improving this project.

Please quote DieMetrics (lite) as follows: See CITATION.md

DiaMetrics (lite) uses Bootstrap v5.3.1 for the design of the website and some visual functions.

DiaMetrics (lite) uses Chart.js v4.4.0 to display the scatter plots.

DiaMetrics (lite) uses the dragdata plugin for Chart.js v2.2.3 to make the separator draggable.

License

Copyright (c) 2024 - 2025 Leon Rosenplänter. DiaMetrics (lite) is available under the MIT license. The full text of the licence can be found in the LICENSE file.

Owner

  • Login: lrosenplaenter
  • Kind: user

Citation (CITATION.cff) 

# This CITATION.cff file was generated with cffinit.
# Visit https://bit.ly/cffinit to generate yours today!

cff-version: 1.2.0
title: DiaMetrics (lite)
message: >-
  If you mention, use, adapt or publish this project
  elsewhere, please cite it using the metadata from this
  file.
type: software
authors:
  - given-names: Leon
    family-names: Rosenplänter
    orcid: 'https://orcid.org/0009-0001-4961-2281'
identifiers:
  - type: doi
    value: 10.5281/zenodo.15002749
    description: Zenodo Archive
repository-code: 'https://github.com/lrosenplaenter/DiaMetrics_lite'
url: 'https://lrosenplaenter.github.io/DiaMetrics_lite'
abstract: >-
  DiaMetrics (lite) is a web-based educational resource for exploring 
  important concepts regarding binary classification (and its 
  evaluation), which is important in many different fields such as 
  psychodiagnostics (e.g. in determining cut-off values for tests), 
  machine learning or medical testing. The lite version is primarily 
  intended for use with direct instruction (e.g. in the context of 
  lectures or talks).
keywords:
  - specificity
  - sensitivity
  - learning-by-doing
  - binary-classification
  - binary-classification-evaluation
  - teaching-materials
  - teaching-resources
  - teaching-tool
  - learning-materials
  - learning-resources
  - learning-tool
license: MIT
commit: v1.1.1
version: 1.1.1
date-released: '2025-03-10'

GitHub Events

Total
  • Push event: 3
  • Create event: 1
Last Year
  • Push event: 3
  • Create event: 1