Public
Edited
Mar 29
Paused
// original algorithm
function spaceSaving(counters, inputStream) {
const e = Array(counters);
const c = Array(counters).fill(0);
const t0 = performance.now();
for (const x of inputStream) {
let i = e.indexOf(x);
if (i === -1) e[(i = d3.leastIndex(c))] = x;
c[i]++;
}
return { elements: e, counts: c, time: Math.floor(performance.now() - t0) };
}
// A variant I'm playing with — maybe 5% faster in some cases
function spaceSavingSorted(k, inputStream) {
const e = Array();
const t0 = performance.now();
const c = Array(k).fill(0);
for (const x of inputStream) {
let i = e.indexOf(x);
if (i === -1) e[(i = Math.min(k - 1, e.length))] = x;
c[i]++;
while (i > 0 && c[i - 1] < c[i]) {
[e[i], e[i - 1]] = [e[i - 1], e[i]];
[c[i], c[i - 1]] = [c[i - 1], c[i]];
i--;
}
}
return {
elements: e,
counts: c,
time: Math.floor(performance.now() - t0)
};
}
spaced = spaceSavingSorted(counters, dataStream)
// ground truth
truth = d3.sort(
d3.rollup(
dataStream,
(l) => l.length,
(d) => d
),
(d) => -d[1]
)
dataStream = {
const r = d3.randomGamma(10);
return Uint8Array.from({ length: 100_000 }, r);
}
import { DuckDBClient_latest as DuckDBClient } from "@mootari/duckdbclient"
db = DuckDBClient.of({ numbers: Array.from(dataStream, (i) => ({ i })) })
db
One platform to build and deploy the best data apps
Experiment and prototype by building visualizations in live JavaScript notebooks. Collaborate with your team and decide which concepts to build out.
Use Observable Framework to build data apps locally. Use data loaders to build in any language or library, including Python, SQL, and R.
Seamlessly deploy to Observable. Test before you ship, use automatic deploy-on-commit, and ensure your projects are always up-to-date.