/
fastmap.js
66 lines (61 loc) · 1.31 KB
/
fastmap.js
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
/**
* FastMap
*/
export default class FastMap {
// http://ibisforest.org/index.php?FastMap
/**
* @param {number} rd Reduced dimension
*/
constructor(rd) {
this._rd = rd
}
/**
* Returns reduced values.
*
* @param {Array<Array<number>>} x Training data
* @returns {Array<Array<number>>} Predicted values
*/
predict(x) {
const d = []
const n = x.length
const m = x[0].length
for (let i = 0; i < n; d[i++] = []);
for (let i = 0; i < n; i++) {
d[i][i] = 0
for (let j = i + 1; j < n; j++) {
let ds = 0
for (let k = 0; k < m; k++) {
ds += (x[i][k] - x[j][k]) ** 2
}
d[i][j] = d[j][i] = Math.sqrt(ds)
}
}
const argmax = arr => {
let idx = -1
let val = -Infinity
for (let i = 0; i < arr.length; i++) {
if (val < arr[i]) {
val = arr[i]
idx = i
}
}
return idx
}
const y = []
for (let i = 0; i < n; y[i++] = []);
for (let k = 0; k < this._rd; k++) {
const p = Math.floor(Math.random() * n)
const a = argmax(d[p])
const b = argmax(d[a])
for (let i = 0; i < n; i++) {
y[i][k] = (d[a][i] ** 2 + d[a][b] ** 2 - d[b][i] ** 2) / (2 * d[a][b])
}
for (let i = 0; i < n; i++) {
for (let j = i + 1; j < n; j++) {
d[i][j] = d[j][i] = Math.sqrt(d[i][j] ** 2 - (y[i][k] - y[j][k]) ** 2)
}
}
}
return y
}
}