w-cluster 1.0.19 → 1.0.21

package/test/kMeans.test.mjs

@@ -0,0 +1,278 @@
+import assert from 'assert'
+import WCluster from '../src/WCluster.mjs'
+
+
+describe(`k-means`, function() {
+
+    let mode = 'k-means'
+    let seed = 0 //給定seed使k-means隨機初始化固定, 分群結果可重現以利穩定測試
+
+    let mat = [
+        [40, 50, 60],
+        [50, 70, 60],
+        [80, 70, 90],
+        [50, 60, 80]
+    ]
+    let clmat = {
+        'keys': null,
+        'ginds': [
+            [0, 1, 3],
+            [2]
+        ],
+        'gmat': [
+            [
+                [-1.704002697669786, 0.43087564048799354],
+                [-0.2509699164590232, -1.1519035967558147],
+                [-0.03633718671228592, 0.48858292291798827]
+            ],
+            [
+                [1.9913098008410954, 0.23244503334983269]
+            ]
+        ],
+        'gltdt': [
+            [
+                [40, 50, 60],
+                [50, 70, 60],
+                [50, 60, 80]
+            ],
+            [
+                [80, 70, 90]
+            ]
+        ]
+    }
+
+    it(`should return '${JSON.stringify(clmat)}' when input ${JSON.stringify(mat)}, { mode, kNumber: 2, nCompNIPALS: 2, seed }`, async function() {
+        let r = await WCluster.cluster(mat, { mode, kNumber: 2, nCompNIPALS: 2, seed })
+        assert.strict.deepStrictEqual(r, clmat)
+    })
+
+    let mat2 = [
+        [1040, 50, 60],
+        [1050, 70, 60],
+        [1080, 70, 90],
+        [1050, 60, 80]
+    ]
+    let clmat2 = {
+        'keys': null,
+        'ginds': [
+            [0, 1, 3],
+            [2]
+        ],
+        'gmat': [
+            [
+                [-1.704002697669786, 0.43087564048799354],
+                [-0.2509699164590232, -1.1519035967558147],
+                [-0.03633718671228592, 0.48858292291798827]
+            ],
+            [
+                [1.9913098008410954, 0.23244503334983269]
+            ]
+        ],
+        'gltdt': [
+            [
+                [1040, 50, 60],
+                [1050, 70, 60],
+                [1050, 60, 80]
+            ],
+            [
+                [1080, 70, 90]
+            ]
+        ]
+    }
+
+    it(`should return '${JSON.stringify(clmat2)}' when input ${JSON.stringify(mat2)}, { mode, kNumber: 2, nCompNIPALS: 2, seed }`, async function() {
+        let r = await WCluster.cluster(mat2, { mode, kNumber: 2, nCompNIPALS: 2, seed })
+        assert.strict.deepStrictEqual(r, clmat2)
+    })
+
+    let mat3 = [
+        [11040, 50, 60],
+        [13050, 70, 60],
+        [15080, 70, 90],
+        [17050, 60, 80]
+    ]
+    let clmat3 = {
+        'keys': null,
+        'ginds': [
+            [0],
+            [1, 2, 3]
+        ],
+        'gmat': [
+            [
+                [-1.8599655569892897, 0.4908764271508211]
+            ],
+            [
+                [-0.3950941652793108, -1.0977355294143445],
+                [1.3430199944041517, -0.17213692267477554],
+                [0.912039727864449, 0.778996024938299]
+            ]
+        ],
+        'gltdt': [
+            [
+                [11040, 50, 60]
+            ],
+            [
+                [13050, 70, 60],
+                [15080, 70, 90],
+                [17050, 60, 80]
+            ]
+        ]
+    }
+
+    it(`should return '${JSON.stringify(clmat3)}' when input ${JSON.stringify(mat3)}, { mode, kNumber: 2, nCompNIPALS: 2, seed }`, async function() {
+        let r = await WCluster.cluster(mat3, { mode, kNumber: 2, nCompNIPALS: 2, seed })
+        assert.strict.deepStrictEqual(r, clmat3)
+    })
+
+    let ltdt = [
+        { name: 'Cameron', a: 40, b: 50, c: 60 },
+        { name: 'Buckley', a: 50, b: 70, c: 60 },
+        { name: 'Paul', a: 80, b: 70, c: 90 },
+        { name: 'Fawcett', a: 50, b: 60, c: 80 },
+    ]
+    let clltdt = {
+        'keys': ['a', 'b', 'c'],
+        'ginds': [
+            [0, 1, 3],
+            [2]
+        ],
+        'gmat': [
+            [
+                [-1.704002697669786, 0.43087564048799354],
+                [-0.2509699164590232, -1.1519035967558147],
+                [-0.03633718671228592, 0.48858292291798827]
+            ],
+            [
+                [1.9913098008410954, 0.23244503334983269]
+            ]
+        ],
+        'gltdt': [
+            [
+                { 'name': 'Cameron', 'a': 40, 'b': 50, 'c': 60 },
+                { 'name': 'Buckley', 'a': 50, 'b': 70, 'c': 60 },
+                { 'name': 'Fawcett', 'a': 50, 'b': 60, 'c': 80 }
+            ],
+            [
+                { 'name': 'Paul', 'a': 80, 'b': 70, 'c': 90 }
+            ]
+        ]
+    }
+
+    it(`should return '${JSON.stringify(clltdt)}' when input ${JSON.stringify(ltdt)}, { mode, kNumber: 2, nCompNIPALS: 2, seed }`, async function() {
+        let r = await WCluster.cluster(ltdt, { mode, kNumber: 2, nCompNIPALS: 2, seed })
+        assert.strict.deepStrictEqual(r, clltdt)
+    })
+
+    it(`should be deterministic (same result on repeated runs) when given seed`, async function() {
+        let r1 = await WCluster.cluster(mat, { mode, kNumber: 2, nCompNIPALS: 2, seed })
+        let r2 = await WCluster.cluster(mat, { mode, kNumber: 2, nCompNIPALS: 2, seed })
+        assert.strict.deepStrictEqual(r1, r2)
+    })
+
+    it(`should not throw and cover all indices when no seed (random init)`, async function() {
+        let r = await WCluster.cluster(mat, { mode, kNumber: 2, nCompNIPALS: 2 })
+        let flat = r.ginds.flat().sort((a, b) => a - b)
+        assert.strict.deepStrictEqual(flat, [0, 1, 2, 3])
+    })
+
+    it(`should return 'Error: data is not eff. array' when input 123`, async function() {
+        let r = ''
+        await WCluster.cluster(123, { mode, seed })
+            .catch((err) => {
+                r = err.toString()
+            })
+        let rr = 'Error: data is not eff. array'
+        assert.strict.deepStrictEqual(r, rr)
+    })
+
+    it(`should return 'Error: data is not eff. array' when input 123.456`, async function() {
+        let r = ''
+        await WCluster.cluster(123.456, { mode, seed })
+            .catch((err) => {
+                r = err.toString()
+            })
+        let rr = 'Error: data is not eff. array'
+        assert.strict.deepStrictEqual(r, rr)
+    })
+
+    it(`should return 'Error: data is not eff. array' when input '123'`, async function() {
+        let r = ''
+        await WCluster.cluster('123', { mode, seed })
+            .catch((err) => {
+                r = err.toString()
+            })
+        let rr = 'Error: data is not eff. array'
+        assert.strict.deepStrictEqual(r, rr)
+    })
+
+    it(`should return 'Error: data is not eff. array' when input '123.456'`, async function() {
+        let r = ''
+        await WCluster.cluster('123.456', { mode, seed })
+            .catch((err) => {
+                r = err.toString()
+            })
+        let rr = 'Error: data is not eff. array'
+        assert.strict.deepStrictEqual(r, rr)
+    })
+
+    it(`should return 'Error: data is not eff. array' when input true`, async function() {
+        let r = ''
+        await WCluster.cluster(true, { mode, seed })
+            .catch((err) => {
+                r = err.toString()
+            })
+        let rr = 'Error: data is not eff. array'
+        assert.strict.deepStrictEqual(r, rr)
+    })
+
+    it(`should return 'Error: data is not eff. array' when input ''`, async function() {
+        let r = ''
+        await WCluster.cluster('', { mode, seed })
+            .catch((err) => {
+                r = err.toString()
+            })
+        let rr = 'Error: data is not eff. array'
+        assert.strict.deepStrictEqual(r, rr)
+    })
+
+    it(`should return 'Error: data is not eff. array' when input []`, async function() {
+        let r = ''
+        await WCluster.cluster([], { mode, seed })
+            .catch((err) => {
+                r = err.toString()
+            })
+        let rr = 'Error: data is not eff. array'
+        assert.strict.deepStrictEqual(r, rr)
+    })
+
+    it(`should return 'Error: data is not eff. array' when input {}`, async function() {
+        let r = ''
+        await WCluster.cluster({}, { mode, seed })
+            .catch((err) => {
+                r = err.toString()
+            })
+        let rr = 'Error: data is not eff. array'
+        assert.strict.deepStrictEqual(r, rr)
+    })
+
+    it(`should return 'Error: data is not eff. array' when input null`, async function() {
+        let r = ''
+        await WCluster.cluster(null, { mode, seed })
+            .catch((err) => {
+                r = err.toString()
+            })
+        let rr = 'Error: data is not eff. array'
+        assert.strict.deepStrictEqual(r, rr)
+    })
+
+    it(`should return 'Error: data is not eff. array' when input undefined`, async function() {
+        let r = ''
+        await WCluster.cluster(undefined, { mode, seed })
+            .catch((err) => {
+                r = err.toString()
+            })
+        let rr = 'Error: data is not eff. array'
+        assert.strict.deepStrictEqual(r, rr)
+    })
+
+})

package/test/kMedoids-large-suggest.test.mjs

@@ -0,0 +1,140 @@
+import assert from 'assert'
+import WCluster from '../src/WCluster.mjs'
+
+
+//確定性PRNG(mulberry32), 使測試資料可重現, 不依賴Math.random
+function mulberry32(a) {
+    return function() {
+        a |= 0
+        a = a + 0x6D2B79F5 | 0
+        let t = Math.imul(a ^ a >>> 15, 1 | a)
+        t = t + Math.imul(t ^ t >>> 7, 61 | t) ^ t
+        return ((t ^ t >>> 14) >>> 0) / 4294967296
+    }
+}
+
+//產生G個分離良好的數值blob(交錯排列, 使各blob點散佈於資料序中), 回傳{data, truth}
+function genBlobs(G, m, D, sep, jit, seed) {
+    let rnd = mulberry32(seed)
+    let centers = []
+    for (let g = 0; g < G; g++) {
+        centers.push(Array.from({ length: D }, (_, d) => g * sep + d * 3))
+    }
+    let data = []
+    let truth = []
+    for (let i = 0; i < m; i++) {
+        for (let g = 0; g < G; g++) {
+            data.push(centers[g].map((c) => c + (rnd() * 2 - 1) * jit))
+            truth.push(g)
+        }
+    }
+    return { data, truth }
+}
+
+//產生G個二元blob(Jaccard場景), 各blob一組base活躍位 + 小幅翻轉
+function genBinaryBlobs(G, m, D, active, flips, seed) {
+    let rnd = mulberry32(seed)
+    let bases = []
+    for (let g = 0; g < G; g++) {
+        let base = new Array(D).fill(0)
+        let s = new Set()
+        while (s.size < active) {
+            s.add(Math.floor(rnd() * D))
+        }
+        s.forEach((i) => { base[i] = 1 })
+        bases.push(base)
+    }
+    let data = []
+    let truth = []
+    for (let i = 0; i < m; i++) {
+        for (let g = 0; g < G; g++) {
+            let v = bases[g].slice()
+            for (let f = 0; f < flips; f++) {
+                let p = Math.floor(rnd() * D)
+                v[p] = v[p] ? 0 : 1
+            }
+            data.push(v)
+            truth.push(g)
+        }
+    }
+    return { data, truth }
+}
+
+//Jaccard距離(0/1陣列)
+let jaccard = (a, b) => {
+    let inter = 0
+    let uni = 0
+    for (let i = 0; i < a.length; i++) {
+        let x = a[i]
+        let y = b[i]
+        if (x & y) inter++
+        if (x | y) uni++
+    }
+    return uni === 0 ? 0 : 1 - inter / uni
+}
+
+//cluster purity: 各分群以多數ground-truth標籤計, 全對則為1
+function purity(ginds, truth) {
+    let correct = 0
+    let n = truth.length
+    for (let g of ginds) {
+        let cnt = {}
+        for (let i of g) {
+            cnt[truth[i]] = (cnt[truth[i]] || 0) + 1
+        }
+        correct += Math.max(...Object.values(cnt))
+    }
+    return correct / n
+}
+
+//flat是否恰好涵蓋0..n-1各一次(不漏不重)
+function coversAll(ginds, n) {
+    let flat = ginds.flat().sort((a, b) => a - b)
+    if (flat.length !== n) {
+        return false
+    }
+    for (let i = 0; i < n; i++) {
+        if (flat[i] !== i) {
+            return false
+        }
+    }
+    return true
+}
+
+
+describe(`k-medoids large suggest`, function() {
+
+    this.timeout(60000)
+
+    it(`should perfectly recover 6 well-separated numeric blobs (n=120, usePCA:false, fasterPAM)`, async function() {
+        let G = 6
+        let m = 20
+        let n = G * m
+        let { data, truth } = genBlobs(G, m, 4, 100, 5, 42)
+        let r = await WCluster.cluster(data, { mode: 'k-medoids', kNumber: G, usePCA: false, useMethod: 'fasterPAM' })
+        assert.strict.strictEqual(coversAll(r.ginds, n), true)
+        assert.strict.strictEqual(r.ginds.length, G)
+        assert.strict.strictEqual(purity(r.ginds, truth), 1)
+    })
+
+    it(`should perfectly recover 6 binary blobs with Jaccard distance (n=120, D=120, usePCA:false, fasterPAM)`, async function() {
+        let G = 6
+        let m = 20
+        let n = G * m
+        let { data, truth } = genBinaryBlobs(G, m, 120, 18, 3, 42)
+        let r = await WCluster.cluster(data, { mode: 'k-medoids', kNumber: G, usePCA: false, funDist: jaccard, useMethod: 'fasterPAM' })
+        assert.strict.strictEqual(coversAll(r.ginds, n), true)
+        assert.strict.strictEqual(r.ginds.length, G)
+        assert.strict.strictEqual(purity(r.ginds, truth), 1)
+    })
+
+    it(`should be deterministic on larger data (same ginds on repeated runs, fasterPAM)`, async function() {
+        let G = 6
+        let m = 20
+        let { data } = genBlobs(G, m, 4, 100, 5, 42)
+        let r1 = await WCluster.cluster(data, { mode: 'k-medoids', kNumber: G, usePCA: false, useMethod: 'fasterPAM' })
+        let r2 = await WCluster.cluster(data, { mode: 'k-medoids', kNumber: G, usePCA: false, useMethod: 'fasterPAM' })
+        assert.strict.deepStrictEqual(r1.ginds, r2.ginds)
+    })
+
+})

package/test/kMedoids-large.test.mjs

@@ -0,0 +1,142 @@
+import assert from 'assert'
+import WCluster from '../src/WCluster.mjs'
+
+
+//確定性PRNG(mulberry32), 使測試資料可重現, 不依賴Math.random
+function mulberry32(a) {
+    return function() {
+        a |= 0
+        a = a + 0x6D2B79F5 | 0
+        let t = Math.imul(a ^ a >>> 15, 1 | a)
+        t = t + Math.imul(t ^ t >>> 7, 61 | t) ^ t
+        return ((t ^ t >>> 14) >>> 0) / 4294967296
+    }
+}
+
+//產生G個分離良好的數值blob(交錯排列, 使各blob點散佈於資料序中), 回傳{data, truth}
+function genBlobs(G, m, D, sep, jit, seed) {
+    let rnd = mulberry32(seed)
+    let centers = []
+    for (let g = 0; g < G; g++) {
+        centers.push(Array.from({ length: D }, (_, d) => g * sep + d * 3))
+    }
+    let data = []
+    let truth = []
+    for (let i = 0; i < m; i++) {
+        for (let g = 0; g < G; g++) {
+            data.push(centers[g].map((c) => c + (rnd() * 2 - 1) * jit))
+            truth.push(g)
+        }
+    }
+    return { data, truth }
+}
+
+//產生G個二元blob(Jaccard場景), 各blob一組base活躍位 + 小幅翻轉
+function genBinaryBlobs(G, m, D, active, flips, seed) {
+    let rnd = mulberry32(seed)
+    let bases = []
+    for (let g = 0; g < G; g++) {
+        let base = new Array(D).fill(0)
+        let s = new Set()
+        while (s.size < active) {
+            s.add(Math.floor(rnd() * D))
+        }
+        s.forEach((i) => {
+            base[i] = 1
+        })
+        bases.push(base)
+    }
+    let data = []
+    let truth = []
+    for (let i = 0; i < m; i++) {
+        for (let g = 0; g < G; g++) {
+            let v = bases[g].slice()
+            for (let f = 0; f < flips; f++) {
+                let p = Math.floor(rnd() * D)
+                v[p] = v[p] ? 0 : 1
+            }
+            data.push(v)
+            truth.push(g)
+        }
+    }
+    return { data, truth }
+}
+
+//Jaccard距離(0/1陣列)
+let jaccard = (a, b) => {
+    let inter = 0
+    let uni = 0
+    for (let i = 0; i < a.length; i++) {
+        let x = a[i]
+        let y = b[i]
+        if (x & y) inter++
+        if (x | y) uni++
+    }
+    return uni === 0 ? 0 : 1 - inter / uni
+}
+
+//cluster purity: 各分群以多數ground-truth標籤計, 全對則為1
+function purity(ginds, truth) {
+    let correct = 0
+    let n = truth.length
+    for (let g of ginds) {
+        let cnt = {}
+        for (let i of g) {
+            cnt[truth[i]] = (cnt[truth[i]] || 0) + 1
+        }
+        correct += Math.max(...Object.values(cnt))
+    }
+    return correct / n
+}
+
+//flat是否恰好涵蓋0..n-1各一次(不漏不重)
+function coversAll(ginds, n) {
+    let flat = ginds.flat().sort((a, b) => a - b)
+    if (flat.length !== n) {
+        return false
+    }
+    for (let i = 0; i < n; i++) {
+        if (flat[i] !== i) {
+            return false
+        }
+    }
+    return true
+}
+
+
+describe(`k-medoids large`, function() {
+
+    this.timeout(60000)
+
+    it(`should perfectly recover 6 well-separated numeric blobs (n=120, usePCA:false)`, async function() {
+        let G = 6
+        let m = 20
+        let n = G * m
+        let { data, truth } = genBlobs(G, m, 4, 100, 5, 42)
+        let r = await WCluster.cluster(data, { mode: 'k-medoids', kNumber: G, usePCA: false, useMethod: 'simple' })
+        assert.strict.strictEqual(coversAll(r.ginds, n), true)
+        assert.strict.strictEqual(r.ginds.length, G)
+        assert.strict.strictEqual(purity(r.ginds, truth), 1)
+    })
+
+    it(`should perfectly recover 6 binary blobs with Jaccard distance (n=120, D=120, usePCA:false)`, async function() {
+        let G = 6
+        let m = 20
+        let n = G * m
+        let { data, truth } = genBinaryBlobs(G, m, 120, 18, 3, 42)
+        let r = await WCluster.cluster(data, { mode: 'k-medoids', kNumber: G, usePCA: false, funDist: jaccard, useMethod: 'simple' })
+        assert.strict.strictEqual(coversAll(r.ginds, n), true)
+        assert.strict.strictEqual(r.ginds.length, G)
+        assert.strict.strictEqual(purity(r.ginds, truth), 1)
+    })
+
+    it(`should be deterministic on larger data (same ginds on repeated runs)`, async function() {
+        let G = 6
+        let m = 20
+        let { data } = genBlobs(G, m, 4, 100, 5, 42)
+        let r1 = await WCluster.cluster(data, { mode: 'k-medoids', kNumber: G, usePCA: false, useMethod: 'simple' })
+        let r2 = await WCluster.cluster(data, { mode: 'k-medoids', kNumber: G, usePCA: false, useMethod: 'simple' })
+        assert.strict.deepStrictEqual(r1.ginds, r2.ginds)
+    })
+
+})

package/test/{cluster.test.mjs → kMedoids-simple.test.mjs}

@@ -2,7 +2,7 @@ import assert from 'assert'
 import WCluster from '../src/WCluster.mjs'
 
 
-describe(`cluster`, function() {
+describe(`k-medoids simple`, function() {
 
     let mode = 'k-medoids'
 
@@ -41,7 +41,7 @@ describe(`cluster`, function() {
     }
 
     it(`should return '${JSON.stringify(clmat)}' when input ${JSON.stringify(mat)}, { mode, kNumber: 2, nCompNIPALS: 2 }`, async function() {
-        let r = await WCluster.cluster(mat, { mode, kNumber: 2, nCompNIPALS: 2 })
+        let r = await WCluster.cluster(mat, { mode, kNumber: 2, nCompNIPALS: 2, useMethod: 'simple' })
         assert.strict.deepStrictEqual(r, clmat)
     })
 
@@ -80,7 +80,7 @@ describe(`cluster`, function() {
     }
 
     it(`should return '${JSON.stringify(clmat2)}' when input ${JSON.stringify(mat2)}, { mode, kNumber: 2, nCompNIPALS: 2 }`, async function() {
-        let r = await WCluster.cluster(mat2, { mode, kNumber: 2, nCompNIPALS: 2 })
+        let r = await WCluster.cluster(mat2, { mode, kNumber: 2, nCompNIPALS: 2, useMethod: 'simple' })
         assert.strict.deepStrictEqual(r, clmat2)
     })
 
@@ -119,7 +119,7 @@ describe(`cluster`, function() {
     }
 
     it(`should return '${JSON.stringify(clmat3)}' when input ${JSON.stringify(mat3)}, { mode, kNumber: 2, nCompNIPALS: 2 }`, async function() {
-        let r = await WCluster.cluster(mat3, { mode, kNumber: 2, nCompNIPALS: 2 })
+        let r = await WCluster.cluster(mat3, { mode, kNumber: 2, nCompNIPALS: 2, useMethod: 'simple' })
         assert.strict.deepStrictEqual(r, clmat3)
     })
 
@@ -158,7 +158,7 @@ describe(`cluster`, function() {
     }
 
     it(`should return '${JSON.stringify(clltdt)}' when input ${JSON.stringify(ltdt)}, { mode, kNumber: 2, nCompNIPALS: 2 }`, async function() {
-        let r = await WCluster.cluster(ltdt, { mode, kNumber: 2, nCompNIPALS: 2 })
+        let r = await WCluster.cluster(ltdt, { mode, kNumber: 2, nCompNIPALS: 2, useMethod: 'simple' })
         assert.strict.deepStrictEqual(r, clltdt)
     })
 

package/toolg/gDocsExams.mjs

@@ -14,7 +14,7 @@ function main() {
     //pks
     let pks = getPks()
 
-    //url
+    //url, urlww
     let url = `https://cdn.jsdelivr.net/npm/w-cluster@${pks.version}/dist/w-cluster.umd.js`
     let urlww = `https://cdn.jsdelivr.net/npm/w-cluster@${pks.version}/dist/w-cluster.wk.umd.js`