Back to metrics

UMAP_Dist

📊 Embedding Quality (UMAP & t-SNE)

UMAP Distance Correlation

Direction

Higher is Better

Value Range

[-1, 1]

Category

Embedding Quality (UMAP & t-SNE)

Visualization quality via coranking analysis

Description

Spearman correlation between latent space and UMAP-reduced pairwise distance matrices

Mathematical Formula

ρ = spearmanr(D_latent.flatten(), D_umap.flatten())

Interpretation Guide

Measures global structure preservation in UMAP visualization. 1 = perfect distance rank preservation.

Related Metrics

Other metrics in the Embedding Quality (UMAP & t-SNE) category

UMAP_Q_local

↑

Average coranking quality for local neighborhoods in UMAP space

UMAP_Q_global

↑

Average coranking quality for global relationships in UMAP space

UMAP_Overall

↑

Comprehensive UMAP quality combining distance correlation, local and global preservation