@thi.ng/distance - npm Package Compare versions

Comparing version 0.2.2 to 0.3.0

CHANGELOG.md

@@ -6,2 +6,14 @@ # Change Log
 
+# [0.3.0](https://github.com/thi-ng/umbrella/compare/@thi.ng/distance@[e-mail]/distance@0.3.0) (2021-04-19)
+
+
+### Features
+
+* **distance:** add argmin*() fns ([72ed376](https://github.com/thi-ng/umbrella/commit/72ed3760c7a6982bcab7d94666957cad90f4f0ef))
+* **distance:** replace HAVERSINE w/ alts ([3a9a77a](https://github.com/thi-ng/umbrella/commit/3a9a77ab0fd06484f2fda5d67c7b151645436a32))
+
+
+
+
+
 ## [0.2.2](https://github.com/thi-ng/umbrella/compare/@thi.ng/distance@[e-mail]/distance@0.2.2) (2021-04-07)
@@ -8,0 +20,0 @@

haversine.d.ts

 import { ReadonlyVec } from "@thi.ng/vectors";
 import { Eucledian } from "./eucledian";
 /**
+ * Distance metric for geo locations given as [lat,lon] vectors.
+ */
+export declare const HAVERSINE_LATLON: Eucledian<ReadonlyVec>;
+/**
  * Distance metric for geo locations given as [lon,lat] vectors.
  */
-export declare const HAVERSINE: Eucledian<ReadonlyVec>;
+export declare const HAVERSINE_LONLAT: Eucledian<ReadonlyVec>;
 //# sourceMappingURL=haversine.d.ts.map

haversine.js

@@ -1,6 +0,10 @@
-import { distHaversine } from "@thi.ng/vectors";
+import { distHaversineLatLon, distHaversineLonLat, } from "@thi.ng/vectors";
 import { Eucledian } from "./eucledian";
 /**
+ * Distance metric for geo locations given as [lat,lon] vectors.
+ */
+export const HAVERSINE_LATLON = new Eucledian(distHaversineLatLon);
+/**
  * Distance metric for geo locations given as [lon,lat] vectors.
  */
-export const HAVERSINE = new Eucledian(distHaversine);
+export const HAVERSINE_LONLAT = new Eucledian(distHaversineLonLat);

index.d.ts

 export * from "./api";
+export * from "./argmin";
 export * from "./eucledian";
 export * from "./haversine";
+export * from "./knearest";
 export * from "./manhattan";
+export * from "./nearest";
 export * from "./squared";
-export * from "./knearest";
-export * from "./nearest";
 //# sourceMappingURL=index.d.ts.map

index.js

 export * from "./api";
+export * from "./argmin";
 export * from "./eucledian";
 export * from "./haversine";
+export * from "./knearest";
 export * from "./manhattan";
+export * from "./nearest";
 export * from "./squared";
-export * from "./knearest";
-export * from "./nearest";

lib/index.js

@@ -5,42 +5,7 @@ 'use strict';
 
-var vectors = require('@thi.ng/vectors');
 var api = require('@thi.ng/api');
 var heaps = require('@thi.ng/heaps');
 var math = require('@thi.ng/math');
+var vectors = require('@thi.ng/vectors');
 
-class Eucledian {
-    constructor(metric) {
-        this.metric = metric;
-    }
-    to(x) {
-        return x;
-    }
-    from(x) {
-        return x;
-    }
-}
-const EUCLEDIAN = new Eucledian(vectors.dist);
-const EUCLEDIAN1 = new Eucledian((a, b) => Math.abs(a - b));
-const EUCLEDIAN2 = new Eucledian(vectors.dist2);
-const EUCLEDIAN3 = new Eucledian(vectors.dist3);
-
-const HAVERSINE = new Eucledian(vectors.distHaversine);
-
-class Manhattan {
-    constructor(dim, metric) {
-        this.dim = dim;
-        this.metric = metric;
-        this._invD = this.dim / Math.sqrt(dim);
-    }
-    to(x) {
-        return x * this._invD;
-    }
-    from(x) {
-        return Math.sqrt((x / this.dim) ** 2 * this.dim);
-    }
-}
-const defManhattan = (dim) => new Manhattan(dim, vectors.distManhattan);
-const MANHATTAN2 = new Manhattan(2, vectors.distManhattan2);
-const MANHATTAN3 = new Manhattan(3, vectors.distManhattan3);
-
 class Squared {
@@ -109,2 +74,60 @@ constructor(metric) {
 
+const argmin = (p, samples, { metric: dist } = DIST_SQ) => {
+    let minD = Infinity;
+    let minArg = -1;
+    for (let i = 0, n = samples.length; i < n; i++) {
+        const d = dist(p, samples[i]);
+        if (d < minD) {
+            minD = d;
+            minArg = i;
+        }
+    }
+    return minArg;
+};
+const argminT = (p, samples, key, dist) => argmin(key(p), samples.map(key), dist);
+const argminK = (k, p, samples, dist = DIST_SQ) => {
+    const neighborhood = knearest(p, k, Infinity, dist);
+    for (let i = 0, n = samples.length; i < n; i++) {
+        neighborhood.consider(samples[i], i);
+    }
+    return neighborhood.values();
+};
+const argminKT = (k, p, samples, key, dist) => argminK(k, key(p), samples.map(key), dist);
+
+class Eucledian {
+    constructor(metric) {
+        this.metric = metric;
+    }
+    to(x) {
+        return x;
+    }
+    from(x) {
+        return x;
+    }
+}
+const EUCLEDIAN = new Eucledian(vectors.dist);
+const EUCLEDIAN1 = new Eucledian((a, b) => Math.abs(a - b));
+const EUCLEDIAN2 = new Eucledian(vectors.dist2);
+const EUCLEDIAN3 = new Eucledian(vectors.dist3);
+
+const HAVERSINE_LATLON = new Eucledian(vectors.distHaversineLatLon);
+const HAVERSINE_LONLAT = new Eucledian(vectors.distHaversineLonLat);
+
+class Manhattan {
+    constructor(dim, metric) {
+        this.dim = dim;
+        this.metric = metric;
+        this._invD = this.dim / Math.sqrt(dim);
+    }
+    to(x) {
+        return x * this._invD;
+    }
+    from(x) {
+        return Math.sqrt((x / this.dim) ** 2 * this.dim);
+    }
+}
+const defManhattan = (dim) => new Manhattan(dim, vectors.distManhattan);
+const MANHATTAN2 = new Manhattan(2, vectors.distManhattan2);
+const MANHATTAN3 = new Manhattan(3, vectors.distManhattan3);
+
 class Nearest {
@@ -153,3 +176,4 @@ constructor(dist, target, radius = Infinity) {
 exports.Eucledian = Eucledian;
-exports.HAVERSINE = HAVERSINE;
+exports.HAVERSINE_LATLON = HAVERSINE_LATLON;
+exports.HAVERSINE_LONLAT = HAVERSINE_LONLAT;
 exports.KNearest = KNearest;
@@ -161,2 +185,6 @@ exports.MANHATTAN2 = MANHATTAN2;
 exports.Squared = Squared;
+exports.argmin = argmin;
+exports.argminK = argminK;
+exports.argminKT = argminKT;
+exports.argminT = argminT;
 exports.defManhattan = defManhattan;
@@ -163,0 +191,0 @@ exports.knearest = knearest;

lib/index.umd.js

@@ -1,1 +0,1 @@
-!function(t,e){"object"==typeof exports&&"undefined"!=typeof module?e(exports,require("@thi.ng/vectors"),require("@thi.ng/api"),require("@thi.ng/heaps"),require("@thi.ng/math")):"function"==typeof define&&define.amd?define(["exports","@thi.ng/vectors","@thi.ng/api","@thi.ng/heaps","@thi.ng/math"],e):e(((t="undefined"!=typeof globalThis?globalThis:t||self).thi=t.thi||{},t.thi.ng=t.thi.ng||{},t.thi.ng.distance={}),t.thi.ng.vectors,t.thi.ng.api,t.thi.ng.heaps,t.thi.ng.math)}(this,(function(t,e,s,i,r){"use strict";class n{constructor(t){this.metric=t}to(t){return t}from(t){return t}}const h=new n(e.dist),a=new n(((t,e)=>Math.abs(t-e))),u=new n(e.dist2),c=new n(e.dist3),d=new n(e.distHaversine);class o{constructor(t,e){this.dim=t,this.metric=e,this._invD=this.dim/Math.sqrt(t)}to(t){return t*this._invD}from(t){return Math.sqrt((t/this.dim)**2*this.dim)}}const l=new o(2,e.distManhattan2),p=new o(3,e.distManhattan3);class m{constructor(t){this.metric=t}to(t){return t*t}from(t){return Math.sqrt(t)}}const f=new m(e.distSq),g=new m(((t,e)=>(t-e)**2)),_=new m(e.distSq2),w=new m(e.distSq3);class v{constructor(t,e,n,h=1/0,a=!1){this.dist=t,this.target=e,this.k=n,this.sorted=a,this._heap=new i.Heap(null,{compare:(t,e)=>e[0]-t[0]}),s.assert(n>0,"invalid k (must be > 0)"),this.radius=r.clamp0(h),this.reset()}reset(){return this._currR=this.dist.to(this.radius),this._heap.clear(),this}deref(){return this.sorted?this._heap.max():this._heap.values}values(){return this.deref().map((t=>t[1]))}includesDistance(t,e=!0){return(e?this.dist.to(t):t)<=this._currR}consider(t,e){const s=this.dist.metric(this.target,t);if(s<=this._currR){const t=this._heap;t.length===this.k?(t.pushPop([s,e]),this._currR=t.peek()[0]):t.push([s,e])}return s}}class N{constructor(t,e,s=1/0){this.dist=t,this.target=e,this.radius=r.clamp0(s),this.reset()}reset(){return this._currR=this.dist.to(this.radius),this.value=void 0,this}deref(){return null!=this.value?[this._currR,this.value]:void 0}includesDistance(t,e=!0){return(e?this.dist.to(t):t)<=this._currR}consider(t,e){const s=this.dist.metric(this.target,t);return s<=this._currR&&(this._currR=s,this.value=e),s}}t.DIST_SQ=f,t.DIST_SQ1=g,t.DIST_SQ2=_,t.DIST_SQ3=w,t.EUCLEDIAN=h,t.EUCLEDIAN1=a,t.EUCLEDIAN2=u,t.EUCLEDIAN3=c,t.Eucledian=n,t.HAVERSINE=d,t.KNearest=v,t.MANHATTAN2=l,t.MANHATTAN3=p,t.Manhattan=o,t.Nearest=N,t.Squared=m,t.defManhattan=t=>new o(t,e.distManhattan),t.knearest=(t,e,s,i=f,r)=>new v(i,t,e,s,r),t.knearest2=(t,e,s,i=_,r)=>new v(i,t,e,s,r),t.knearest3=(t,e,s,i=w,r)=>new v(i,t,e,s,r),t.knearestN=(t,e,s,i=g,r)=>new v(i,t,e,s,r),t.nearest=(t,e,s=f)=>new N(s,t,e),t.nearest2=(t,e,s=_)=>new N(s,t,e),t.nearest3=(t,e,s=w)=>new N(s,t,e),t.nearestN=(t,e,s=g)=>new N(s,t,e),Object.defineProperty(t,"__esModule",{value:!0})}));
+!function(t,e){"object"==typeof exports&&"undefined"!=typeof module?e(exports,require("@thi.ng/api"),require("@thi.ng/heaps"),require("@thi.ng/math"),require("@thi.ng/vectors")):"function"==typeof define&&define.amd?define(["exports","@thi.ng/api","@thi.ng/heaps","@thi.ng/math","@thi.ng/vectors"],e):e(((t="undefined"!=typeof globalThis?globalThis:t||self).thi=t.thi||{},t.thi.ng=t.thi.ng||{},t.thi.ng.distance={}),t.thi.ng.api,t.thi.ng.heaps,t.thi.ng.math,t.thi.ng.vectors)}(this,(function(t,e,s,i,r){"use strict";class n{constructor(t){this.metric=t}to(t){return t*t}from(t){return Math.sqrt(t)}}const h=new n(r.distSq),a=new n(((t,e)=>(t-e)**2)),u=new n(r.distSq2),c=new n(r.distSq3);class o{constructor(t,r,n,h=1/0,a=!1){this.dist=t,this.target=r,this.k=n,this.sorted=a,this._heap=new s.Heap(null,{compare:(t,e)=>e[0]-t[0]}),e.assert(n>0,"invalid k (must be > 0)"),this.radius=i.clamp0(h),this.reset()}reset(){return this._currR=this.dist.to(this.radius),this._heap.clear(),this}deref(){return this.sorted?this._heap.max():this._heap.values}values(){return this.deref().map((t=>t[1]))}includesDistance(t,e=!0){return(e?this.dist.to(t):t)<=this._currR}consider(t,e){const s=this.dist.metric(this.target,t);if(s<=this._currR){const t=this._heap;t.length===this.k?(t.pushPop([s,e]),this._currR=t.peek()[0]):t.push([s,e])}return s}}const d=(t,e,s,i=h,r)=>new o(i,t,e,s,r),l=(t,e,{metric:s}=h)=>{let i=1/0,r=-1;for(let n=0,h=e.length;n<h;n++){const h=s(t,e[n]);h<i&&(i=h,r=n)}return r},m=(t,e,s,i=h)=>{const r=d(e,t,1/0,i);for(let t=0,e=s.length;t<e;t++)r.consider(s[t],t);return r.values()};class p{constructor(t){this.metric=t}to(t){return t}from(t){return t}}const g=new p(r.dist),f=new p(((t,e)=>Math.abs(t-e))),_=new p(r.dist2),w=new p(r.dist3),v=new p(r.distHaversineLatLon),N=new p(r.distHaversineLonLat);class A{constructor(t,e){this.dim=t,this.metric=e,this._invD=this.dim/Math.sqrt(t)}to(t){return t*this._invD}from(t){return Math.sqrt((t/this.dim)**2*this.dim)}}const S=new A(2,r.distManhattan2),T=new A(3,r.distManhattan3);class E{constructor(t,e,s=1/0){this.dist=t,this.target=e,this.radius=i.clamp0(s),this.reset()}reset(){return this._currR=this.dist.to(this.radius),this.value=void 0,this}deref(){return null!=this.value?[this._currR,this.value]:void 0}includesDistance(t,e=!0){return(e?this.dist.to(t):t)<=this._currR}consider(t,e){const s=this.dist.metric(this.target,t);return s<=this._currR&&(this._currR=s,this.value=e),s}}t.DIST_SQ=h,t.DIST_SQ1=a,t.DIST_SQ2=u,t.DIST_SQ3=c,t.EUCLEDIAN=g,t.EUCLEDIAN1=f,t.EUCLEDIAN2=_,t.EUCLEDIAN3=w,t.Eucledian=p,t.HAVERSINE_LATLON=v,t.HAVERSINE_LONLAT=N,t.KNearest=o,t.MANHATTAN2=S,t.MANHATTAN3=T,t.Manhattan=A,t.Nearest=E,t.Squared=n,t.argmin=l,t.argminK=m,t.argminKT=(t,e,s,i,r)=>m(t,i(e),s.map(i),r),t.argminT=(t,e,s,i)=>l(s(t),e.map(s),i),t.defManhattan=t=>new A(t,r.distManhattan),t.knearest=d,t.knearest2=(t,e,s,i=u,r)=>new o(i,t,e,s,r),t.knearest3=(t,e,s,i=c,r)=>new o(i,t,e,s,r),t.knearestN=(t,e,s,i=a,r)=>new o(i,t,e,s,r),t.nearest=(t,e,s=h)=>new E(s,t,e),t.nearest2=(t,e,s=u)=>new E(s,t,e),t.nearest3=(t,e,s=c)=>new E(s,t,e),t.nearestN=(t,e,s=a)=>new E(s,t,e),Object.defineProperty(t,"__esModule",{value:!0})}));

package.json

 {
     "name": "@thi.ng/distance",
-    "version": "0.2.2",
+    "version": "0.3.0",
     "description": "N-dimensional distance metrics & K-nearest neighborhoods for point queries",
@@ -55,3 +55,3 @@ "module": "./index.js",
         "@thi.ng/math": "^3.4.0",
-        "@thi.ng/vectors": "^5.2.2"
+        "@thi.ng/vectors": "^5.3.0"
     },
@@ -87,3 +87,3 @@ "files": [
     },
-    "gitHead": "9d35fc75c08bcfdcd67126b8183c869fda71573a"
+    "gitHead": "a099b0a1f1c6180819df9cc4786faa5c2b535140"
 }

README.md

@@ -36,16 +36,17 @@ <!-- This file is generated - DO NOT EDIT! -->
 
-| **Preset**        | **Number** | **n-D** | **2D** | **3D** | **Comments**                                                         |
-|-------------------|------------|---------|--------|--------|----------------------------------------------------------------------|
-| `EUCLEDIAN`       |            | ✅       |        |        | Eucledian distance                                                   |
-| `EUCLEDIAN1`      | ✅          |         |        |        |                                                                      |
-| `EUCLEDIAN2`      |            |         | ✅      |        |                                                                      |
-| `EUCLEDIAN3`      |            |         |        | ✅      |                                                                      |
-| `HAVERSINE`       |            |         | ✅      |        | Great-circle distance for lon/lat geo locations                      |
-| `DIST_SQ`         |            | ✅       |        |        | Squared dist (avoids `Math.sqrt`)                                    |
-| `DIST_SQ1`        | ✅          |         |        |        |                                                                      |
-| `DIST_SQ2`        |            |         | ✅      |        |                                                                      |
-| `DIST_SQ3`        |            |         |        | ✅      |                                                                      |
-| `defManhattan(n)` |            | ✅       |        |        | [Manhattan distance](https://en.wikipedia.org/wiki/Taxicab_geometry) |
-| `MANHATTAN2`      |            |         | ✅      |        |                                                                      |
-| `MANHATTAN3`      |            |         |        | ✅      |                                                                      |
+| **Preset**         | **Number** | **nD** | **2D** | **3D** | **Comments**                                                         |
+|--------------------|------------|--------|--------|--------|----------------------------------------------------------------------|
+| `EUCLEDIAN`        |            | ✅      |        |        | Eucledian distance                                                   |
+| `EUCLEDIAN1`       | ✅          |        |        |        |                                                                      |
+| `EUCLEDIAN2`       |            |        | ✅      |        |                                                                      |
+| `EUCLEDIAN3`       |            |        |        | ✅      |                                                                      |
+| `HAVERSINE_LATLON` |            |        | ✅      |        | Great-circle distance for lat/lon geo locations                      |
+| `HAVERSINE_LONLAT` |            |        | ✅      |        | Great-circle distance for lon/lat geo locations                      |
+| `DIST_SQ`          |            | ✅      |        |        | Squared dist (avoids `Math.sqrt`)                                    |
+| `DIST_SQ1`         | ✅          |        |        |        |                                                                      |
+| `DIST_SQ2`         |            |        | ✅      |        |                                                                      |
+| `DIST_SQ3`         |            |        |        | ✅      |                                                                      |
+| `defManhattan(n)`  |            | ✅      |        |        | [Manhattan distance](https://en.wikipedia.org/wiki/Taxicab_geometry) |
+| `MANHATTAN2`       |            |        | ✅      |        |                                                                      |
+| `MANHATTAN3`       |            |        |        | ✅      |                                                                      |
 
@@ -111,3 +112,3 @@ ### Neighborhoods
 
-Package sizes (gzipped, pre-treeshake): ESM: 887 bytes / CJS: 994 bytes / UMD: 1.01 KB
+Package sizes (gzipped, pre-treeshake): ESM: 1.06 KB / CJS: 1.17 KB / UMD: 1.21 KB
 
@@ -114,0 +115,0 @@ ## Dependencies

No alert changes

Improved metrics

Total package byte prevSize

72198

increased by13.29%

Number of package files

26

increased by8.33%

Lines of code

969

increased by19.04%

Number of lines in readme file

170

increased by0.59%

Dependency changes

• [e-mail]/vectors@^5.3.0