KDE estimation / Fil

Fil

Vocateur.

Workspace

Public

Edited

Nov 21, 2024

2 forks

6 stars

Algorithms

Hello, A5 Chandrupatla’s root-finding method Sidi’s root-finding method Regular numbers DruidJS worker Natural breaks Distance to a segment Ray out of a convex hull Word Tour: 40k words and their friends Hello, @thi.ng/grid-iterators Head/tail breaks Pseudo-blue noise shader How fast does walk-on-spheres converge?AoC 12: shortest path under constraints

KDE estimation

Plot: Correlation heatmap Poisson Finish 2 Poisson disk sampling functions WoS with transport Simple and surprising sort Local median Time series topological subsampling Union-Find Level set experiment 1 Mean value coordinates Poisson potential Middle-square World of squares (spherical)World of squares Largest Inscribed Square Hello, PyWavelets Geothmetic meandian Hello, Reorder.js Geometric Median Image FFT Transport to a map Disc Transport TP3: Power Diagram and Semi-Discrete Optimal Transport The blue wave Hello, genetic-js Sliced Optimal Transport DruidJS Self-Organizing Maps meet Delaunay Hello, polygon-clipping seedrandom, minimal Walk on Spheres 2 Walk on Spheres Hello, Autoencoder Kaprekar’s number VoronoiMap2D Hello, ccwt.js Polygon Triangulation Quantile.invert?Linear congruential generator Hue blur Needle in a haystack Moving average blur Apollo 11 implementation of trigonometric functions, by Margaret H. Hamilton (march 1969)2D curves intersection The 2D approximate Newton-Raphson method Inverting Lee’s Tetrahedral projection Linde–Buzo–Gray stippling Mean shift clustering with kd-tree 2D point distributions Shortest path Kahan Summation Hello, delatin Dijkstra’s algorithm in gpu.js Lloyd’s relaxation on a graph Manhattan Diameter Manhattan Voronoi Mobility landscapes — an introduction Dijkstra’s shortest-path tree H3 oddities Protein Matrix Convex Spectral Weights Sort stuff by similarity Kriging Delaunay.findTriangle n-dimensions binning?Travelling with a self-organizing map UMAP-o-Matic MNIST & UMAP-js Hello UMAP-js Mean shift clustering Levenshtein transition Rd quasi-random sequences Automated label placement (countries)Phyllotaxis explained Motionrugs Planar hull (Andrew’s monotone chain algorithm)South Africa’s medial axis Travelling salesperson approximation with t-SNE Distance to shore Worker ngraph: pagerank, louvain…t-SNE Voronoi Cloud Contours Circular function drawing Kruskal Maze Mycelium Travelling salesperson approximation on the globe, with t-SNE tsne.js tsne.js & worker

Also listed in…

Interpolation

k = {

const t0 = performance.now();

const W = 1200;

const density2d = kde.density2d(points, {

bandwidth,

extent: [

[0, W],

[0, W]

bins: [W, W]

});

const canvas = DOM.canvas(W, W);

canvas.style.width = `${width / 2 - 10}px`;

canvas.style.height = `${width / 2 - 10}px`;

density2d.heatmap({ canvas, color });

const time = performance.now() - t0;

return Object.assign(canvas, { time });

}

b = {

const t0 = performance.now();

const W = 1200;

const data = new Float64Array(W * W);

for (const p of points) data[Math.floor(p[0]) + W * Math.floor(W - p[1])]++;

d3.blur2({ data, width: W }, bandwidth);

const canvas = heatmap(data, {

width: W,

upscale: (width / 2 - 10) / W,

color

});

const time = performance.now() - t0;

return Object.assign(canvas, { time });

}

kde = require("fast-kde@0.2")

k.time

b.time

import { heatmap } from "@fil/heatmap"

color = d3.scaleLinear().domain([0, 1]).range(["#000000", "#FFFFFF"])

Purpose-built for displays of data

Observable is your go-to platform for exploring data and creating expressive data visualizations. Use reactive JavaScript notebooks for prototyping and a collaborative canvas for visual data exploration and dashboard creation.

Learn more