Applying GIS, Data Vis. and Deep Learning to environmental science and nuclear emergency responses
Public
Edited
Aug 4, 2023
1 star
// https://xunli.gitbook.io/jsgeoda/spatial-weights/distance-based-weights
// https://observablehq.com/d/eb57e37649725f96
// geoda = jsgeoda.New()
// encoder = new TextEncoder()
// arrayBuffer = encoder.encode(JSON.stringify(toGeoJSON(data)))
// dataDa = geoda.read_geojson(arrayBuffer)
// act = geoda.get_col(dataDa, "activity")
// geoda.get_numobs(dataDa)
// weights = geoda.getKnnWeights(dataDa, 10)
// redcap = geoda.skater(weights, 10, [act])
// new Set(redcap.clusters)
// activityCluster = Plot.plot({
// // color: {
// // type: "threshold",
// // scheme: "orrd",
// // domain: [0.25, 0.5, 0.75, 0.95, 0.99].map((d) => d3.quantile(activity, d))
// // },
// projection: {
// type: "orthographic",
// rotate: [30, -40]
// },
// marks: [
// Plot.geo(land, { fill: "#333", stroke: "black", strokeWidth: 0.5 }),
// Plot.graticule(),
// Plot.dot(data, {
// x: "lon",
// y: "lat",
// fill: (d, i) => redcap.clusters[i],
// r: 1
// }),
// Plot.sphere({ strokeWidth: 0.5 })
// ]
// })
activity_norm = norm(data.map((d) => d.activity))
lon_norm = norm(data.map((d) => d.lon))
lat_norm = norm(data.map((d) => d.lat))
time_norm = norm(data.map((d) => d.begperiod.getTime()))
points = Array.from({ length: data.length }).map((_, i) => [
(lon_norm[i] * w1) / 2,
(lat_norm[i] * w1) / 2,
(w2 * (activity_norm[i] * (1 - w1))) / 2,
((1 - w2) * (time_norm[i] * (1 - w1))) / 2
])
activity = data.map((d) => d.activity)
world = (
await fetch("https://cdn.jsdelivr.net/npm/world-atlas@2/land-50m.json")
).json()
land = topojson.feature(world, world.objects.land)
function toGeoJSON(data) {
const features = data.map((d) => ({
type: "feature",
geometry: {
type: "Point",
coordinates: [d.lon, d.lat]
},
properties: {
area_id: d.area_id,
depth: d.depth,
time: d.begperiod,
activity: d.activity
}
}));
return { type: "FeatureCollection", features: features };
}
// jsgeoda = import("https://cdn.skypack.dev/jsgeoda")
// import { meanshift, kernel, distance } from "@fil/mean-shift-clustering"
// kmeans = require("https://bundle.run/kmeans-engine@1.5.0")
// import("https://cdn.skypack.dev/density-clustering@1.3.0?min")
// https://github.com/uhho/density-clustering
clustering = require("https://bundle.run/density-clustering@1.3.0")
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.
