AICGijon/quantificationlib

QuantificationLib is an open-source library for quantification learning.

36
/ 100
Emerging

QuantificationLib helps data scientists and machine learning engineers accurately estimate the proportion of positive or negative instances within an unlabeled dataset, even when the models used to classify individual items are imperfect. It takes an existing dataset of classified items and improves the accuracy of the overall class distribution count. This is useful for anyone needing to understand population-level trends rather than individual predictions.

No commits in the last 6 months. Available on PyPI.

Use this if you need to know 'how many' items of a certain type are in a large collection, rather than 'which specific' items are of that type.

Not ideal if your primary goal is to improve the accuracy of individual item classifications, rather than the overall class distribution.

data-science machine-learning-engineering population-analysis class-proportion-estimation unlabeled-data-analysis
Stale 6m No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 25 / 25
Community 6 / 25

How are scores calculated?

Stars

12

Forks

1

Language

Python

License

GPL-3.0

Last pushed

Apr 06, 2024

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/ml-frameworks/AICGijon/quantificationlib"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.