Stuck on an issue?

Lightrun Answers was designed to reduce the constant googling that comes with debugging 3rd party libraries. It collects links to all the places you might be looking at while hunting down a tough bug.

And, if you’re still stuck at the end, we’re happy to hop on a call to see how we can help out.

Feature request: phik correlation metrics

See original GitHub issue

It’d be great to add other correlations as distances, such as Spearman’s rho or the new correlation coefficient called phi_k. I am aware that it’s possible to use any distance metric using either metric='precomputed' or implementing a custom function, but it’d be more convenient to pass metric='spearmanrho' or 'phik' directly.

Issue Analytics

State:
Created 4 years ago
Comments:5 (1 by maintainers)

Top GitHub Comments

1reaction

sleighsoftcommented, Nov 20, 2019

Added a PR.

0reactions

sleighsoftcommented, Nov 20, 2019

@gokceneraslan I look at this again and I believe the best place for this would actually be the pynndescent project https://github.com/lmcinnes/pynndescent/blob/master/pynndescent/distances.py

It has the same number of metrics as UMAP and I assume it will be the default for UMAP.

Top Results From Across the Web

Phik (𝜙k) — get familiar with the latest correlation coefficient

Phik ( 𝜙 k) is a new and practical correlation coefficient that works consistently between categorical, ordinal and interval variables, ...

How to Choose a Feature Selection Method For Machine ...

Filter-based feature selection methods use statistical measures to score the correlation or dependence between input variables that can be ...

phik package — Phi_K correlation library documentation

Description: Convert Pearson correlation value into a chi2 value of a contingency test matrix of a bivariate gaussion, and vice-versa.

A new correlation coefficient between categorical, ordinal and ...

Request PDF | A new correlation coefficient between categorical, ... 1 shows the correlation between the features according to Phik coefficient [20] ....

Finding out the top 10 corr features in a data in R

If we handle the cor() output as an adjacency matrix for a network object, igraph can help us to transform the output to...