Merge pull request #15 from Kanaries/codex/complete-missing-algorithms-in-docs

Add missing algorithm docs

Author: ObservedObserver

docs/_sidebar.md

@@ -4,9 +4,12 @@
     + [Algebra(代数)](zh-cn/apis/algebra/index.md)
     + [Clusters](zh-cn/apis/clusters/index.md)
         + [KMeans](zh-cn/apis/clusters/kmeans.md)
+        + [DBScan](zh-cn/apis/clusters/dbscan.md)
     + [Ensemble](zh-cn/apis/ensemble.md)
         + [IsolationForest(孤立森林)](zh-cn/apis/ensemble/iforest.md)
     + [Neighbors](zh-cn/apis/neighbors.md)
         + [KNearstNeighbors(K近邻)](zh-cn/apis/neighbors/knn.md)
     + [Tree](zh-cn/apis/tree/index.md)
-        + [DecisionTreeClassifier(决策树分类器)](zh-cn/apis/tree/decisionTreeClassifier.md)
+        + [DecisionTreeClassifier(决策树分类器)](zh-cn/apis/tree/decisionTreeClassifier.md)
+    + [Linear](zh-cn/apis/linear/index.md)
+        + [LinearRegression(线性回归)](zh-cn/apis/linear/linearRegression.md)

docs/en/_sidebar.md

@@ -1,4 +1,6 @@
 + APIs
     + [Clusters](./apis/clusters.md)
     + [Ensemble](./apis/ensemble.md)
-    + [Neighbors](./apis/neighbors.md)
+    + [Neighbors](./apis/neighbors.md)
+    + [Tree](./apis/tree/index.md)
+    + [Linear](./apis/linear/index.md)

docs/en/apis/clusters.md

@@ -28,4 +28,18 @@ const kmeans = new KMeans(2, 0.05, initCenters);
 
 const result = kmeans.fitPredict(X, sampleWeights);
 
-```
+```
+
+## Clusters.DBScan
+
+```ts
+constructor(eps: number = 0.5, minSamples: number = 5, distanceType: Distance.IDistanceType = 'euclidiean')
+```
+
+`fitPredict(samplesX: number[][]): number[]` returns cluster labels for samples. Noise points are marked as `-1`.
+
+```ts
+const X = makeCircles(20, 20, 1, 5);
+const dbscan = new DBScan(0.6, 3);
+const labels = dbscan.fitPredict(X);
+```

docs/en/apis/ensemble.md

@@ -0,0 +1,5 @@
+# Ensemble
+
+## Ensemble.IsolationForest
+
+See [ensemble/iforest.md](ensemble/iforest.md) for details.

docs/en/apis/ensemble/iforest.md

@@ -0,0 +1,17 @@
+## IsolationForest
+
+```ts
+constructor(subsampling_size: number = 256, tree_num: number = 100, contamination: 'auto' | number = 'auto')
+```
+
+### Methods
++ `fit(samplesX: number[][]): void`
++ `predict(samplesX: number[][]): (0|1)[]`
+
+### Example
+```ts
+const iForest = new IsolationForest(256, 10, 0.25);
+const X = [[-2, -1], [-1, -1], [-1, -2], [1, 1]];
+iForest.fit(X);
+const result = iForest.predict(X);
+```

docs/en/apis/linear/index.md

@@ -0,0 +1,3 @@
+# Linear
+
+- [LinearRegression](linearRegression.md)

docs/en/apis/linear/linearRegression.md

@@ -0,0 +1,16 @@
+## LinearRegression
+
+```ts
+constructor()
+```
+
+### Methods
++ `fit(X: number[][], Y: number[]): void`
++ `predict(X: number[][]): number[]`
+
+### Example
+```ts
+const lr = new LinearRegression();
+lr.fit([[0], [1]], [1, 3]);
+const pred = lr.predict([[2]]); // about 5
+```

docs/en/apis/neighbors.md

@@ -0,0 +1,3 @@
+# Neighbors
+
+- [KNearstNeighbors](neighbors/knn.md)

docs/en/apis/neighbors/knn.md

@@ -0,0 +1,17 @@
+## KNearstNeighbors
+
+```ts
+constructor(
+    kNeighbors: number = 5,
+    weightType: IWeightType = 'uniform',
+    distanceType: Distance.IDistanceType = 'euclidiean',
+    pNorm: number = 2
+)
+```
+
+### Example
+```ts
+const knn = new KNearstNeighbors(3, 'distance', '2-norm');
+knn.fit(trainX, trainY);
+const result = knn.predict(trainX);
+```

docs/en/apis/tree/decisionTreeClassifier.md

@@ -0,0 +1,18 @@
+## DecisionTreeClassifier
+
+```ts
+interface DecisionTreeProps {
+    max_depth?: number;
+    min_samples_split?: number;
+    criterion?: 'entropy' | 'gini';
+}
+
+constructor(props: DecisionTreeProps = {})
+```
+
+### Example
+```ts
+const dt = new DecisionTreeClassifier({ criterion: 'gini' });
+dt.fit(X, Y);
+const result = dt.predict(T);
+```