nlptoolkit-classification 1.0.0

package/dist/Classifier/Classifier.js

@@ -0,0 +1,72 @@
+(function (factory) {
+    if (typeof module === "object" && typeof module.exports === "object") {
+        var v = factory(require, exports);
+        if (v !== undefined) module.exports = v;
+    }
+    else if (typeof define === "function" && define.amd) {
+        define(["require", "exports", "../Attribute/DiscreteAttribute", "../Attribute/DiscreteIndexedAttribute", "../Performance/ConfusionMatrix", "../Performance/DetailedClassificationPerformance"], factory);
+    }
+})(function (require, exports) {
+    "use strict";
+    Object.defineProperty(exports, "__esModule", { value: true });
+    exports.Classifier = void 0;
+    const DiscreteAttribute_1 = require("../Attribute/DiscreteAttribute");
+    const DiscreteIndexedAttribute_1 = require("../Attribute/DiscreteIndexedAttribute");
+    const ConfusionMatrix_1 = require("../Performance/ConfusionMatrix");
+    const DetailedClassificationPerformance_1 = require("../Performance/DetailedClassificationPerformance");
+    class Classifier {
+        /**
+         * Checks given instance's attribute and returns true if it is a discrete indexed attribute, false otherwise.
+         *
+         * @param instance Instance to check.
+         * @return True if instance is a discrete indexed attribute, false otherwise.
+         */
+        discreteCheck(instance) {
+            for (let i = 0; i < instance.attributeSize(); i++) {
+                if (instance.getAttribute(i) instanceof DiscreteAttribute_1.DiscreteAttribute &&
+                    !(instance.getAttribute(i) instanceof DiscreteIndexedAttribute_1.DiscreteIndexedAttribute)) {
+                    return false;
+                }
+            }
+            return true;
+        }
+        /**
+         * TestClassification an instance list with the current model.
+         *
+         * @param testSet Test data (list of instances) to be tested.
+         * @return The accuracy (and error) of the model as an instance of Performance class.
+         */
+        test(testSet) {
+            let classLabels = testSet.getUnionOfPossibleClassLabels();
+            let confusion = new ConfusionMatrix_1.ConfusionMatrix(classLabels);
+            for (let i = 0; i < testSet.size(); i++) {
+                let instance = testSet.get(i);
+                confusion.classify(instance.getClassLabel(), this.model.predict(instance));
+            }
+            return new DetailedClassificationPerformance_1.DetailedClassificationPerformance(confusion);
+        }
+        /**
+         * Runs current classifier with the given train and test data.
+         *
+         * @param parameter Parameter of the classifier to be trained.
+         * @param trainSet  Training data to be used in training the classifier.
+         * @param testSet   Test data to be tested after training the model.
+         * @return The accuracy (and error) of the trained model as an instance of Performance class.
+         * @throws DiscreteFeaturesNotAllowed Exception for discrete features.
+         */
+        singleRun(parameter, trainSet, testSet) {
+            this.train(trainSet, parameter);
+            return this.test(testSet);
+        }
+        /**
+         * Accessor for the model.
+         *
+         * @return Model.
+         */
+        getModel() {
+            return this.model;
+        }
+    }
+    exports.Classifier = Classifier;
+});
+//# sourceMappingURL=Classifier.js.map

package/dist/Classifier/Classifier.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"Classifier.js","sourceRoot":"","sources":["../../source/Classifier/Classifier.ts"],"names":[],"mappings":";;;;;;;;;;;;IAIA,sEAAiE;IACjE,oFAA+E;IAC/E,oEAA+D;IAE/D,wGAAmG;IAEnG,MAAsB,UAAU;QAM5B;;;;;WAKG;QACH,aAAa,CAAC,QAAkB;YAC5B,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,QAAQ,CAAC,aAAa,EAAE,EAAE,CAAC,EAAE,EAAE;gBAC/C,IAAI,QAAQ,CAAC,YAAY,CAAC,CAAC,CAAC,YAAY,qCAAiB;oBACrD,CAAC,CAAC,QAAQ,CAAC,YAAY,CAAC,CAAC,CAAC,YAAY,mDAAwB,CAAC,EAAE;oBACjE,OAAO,KAAK,CAAC;iBAChB;aACJ;YACD,OAAO,IAAI,CAAC;QAChB,CAAC;QAED;;;;;WAKG;QACH,IAAI,CAAC,OAAqB;YACtB,IAAI,WAAW,GAAG,OAAO,CAAC,6BAA6B,EAAE,CAAC;YAC1D,IAAI,SAAS,GAAG,IAAI,iCAAe,CAAC,WAAW,CAAC,CAAC;YACjD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,OAAO,CAAC,IAAI,EAAE,EAAE,CAAC,EAAE,EAAE;gBACrC,IAAI,QAAQ,GAAG,OAAO,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC;gBAC9B,SAAS,CAAC,QAAQ,CAAC,QAAQ,CAAC,aAAa,EAAE,EAAE,IAAI,CAAC,KAAK,CAAC,OAAO,CAAC,QAAQ,CAAC,CAAC,CAAC;aAC9E;YACD,OAAO,IAAI,qEAAiC,CAAC,SAAS,CAAC,CAAC;QAC5D,CAAC;QAED;;;;;;;;WAQG;QACH,SAAS,CAAC,SAAoB,EAAE,QAAsB,EAAE,OAAqB;YACzE,IAAI,CAAC,KAAK,CAAC,QAAQ,EAAE,SAAS,CAAC,CAAC;YAChC,OAAO,IAAI,CAAC,IAAI,CAAC,OAAO,CAAC,CAAC;QAC9B,CAAC;QAED;;;;WAIG;QACH,QAAQ;YACJ,OAAO,IAAI,CAAC,KAAK,CAAA;QACrB,CAAC;KACJ;IA5DD,gCA4DC"}

package/dist/Classifier/DeepNetwork.d.ts

@@ -0,0 +1,13 @@
+import { Classifier } from "./Classifier";
+import { InstanceList } from "../InstanceList/InstanceList";
+import { Parameter } from "../Parameter/Parameter";
+export declare class DeepNetwork extends Classifier {
+    /**
+     * Training algorithm for deep network classifier.
+     *
+     * @param trainSet   Training data given to the algorithm.
+     * @param parameters Parameters of the deep network algorithm. crossValidationRatio and seed are used as parameters.
+     * @throws DiscreteFeaturesNotAllowed Exception for discrete features.
+     */
+    train(trainSet: InstanceList, parameters: Parameter): void;
+}

package/dist/Classifier/DeepNetwork.js

@@ -0,0 +1,31 @@
+(function (factory) {
+    if (typeof module === "object" && typeof module.exports === "object") {
+        var v = factory(require, exports);
+        if (v !== undefined) module.exports = v;
+    }
+    else if (typeof define === "function" && define.amd) {
+        define(["require", "exports", "./Classifier", "../InstanceList/Partition", "../Model/DeepNetworkModel"], factory);
+    }
+})(function (require, exports) {
+    "use strict";
+    Object.defineProperty(exports, "__esModule", { value: true });
+    exports.DeepNetwork = void 0;
+    const Classifier_1 = require("./Classifier");
+    const Partition_1 = require("../InstanceList/Partition");
+    const DeepNetworkModel_1 = require("../Model/DeepNetworkModel");
+    class DeepNetwork extends Classifier_1.Classifier {
+        /**
+         * Training algorithm for deep network classifier.
+         *
+         * @param trainSet   Training data given to the algorithm.
+         * @param parameters Parameters of the deep network algorithm. crossValidationRatio and seed are used as parameters.
+         * @throws DiscreteFeaturesNotAllowed Exception for discrete features.
+         */
+        train(trainSet, parameters) {
+            let partition = new Partition_1.Partition(trainSet, parameters.getCrossValidationRatio(), true);
+            this.model = new DeepNetworkModel_1.DeepNetworkModel(partition.get(1), partition.get(0), parameters);
+        }
+    }
+    exports.DeepNetwork = DeepNetwork;
+});
+//# sourceMappingURL=DeepNetwork.js.map

package/dist/Classifier/DeepNetwork.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"DeepNetwork.js","sourceRoot":"","sources":["../../source/Classifier/DeepNetwork.ts"],"names":[],"mappings":";;;;;;;;;;;;IAAA,6CAAwC;IAGxC,yDAAoD;IAEpD,gEAA2D;IAE3D,MAAa,WAAY,SAAQ,uBAAU;QAEvC;;;;;;WAMG;QACH,KAAK,CAAC,QAAsB,EAAE,UAAqB;YAC/C,IAAI,SAAS,GAAG,IAAI,qBAAS,CAAC,QAAQ,EAA0B,UAAW,CAAC,uBAAuB,EAAE,EAAE,IAAI,CAAC,CAAC;YAC7G,IAAI,CAAC,KAAK,GAAG,IAAI,mCAAgB,CAAC,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,EAAyB,UAAU,CAAC,CAAC;QAC7G,CAAC;KAEJ;IAdD,kCAcC"}

package/dist/Classifier/Dummy.d.ts

@@ -0,0 +1,13 @@
+import { Classifier } from "./Classifier";
+import { InstanceList } from "../InstanceList/InstanceList";
+import { Parameter } from "../Parameter/Parameter";
+export declare class Dummy extends Classifier {
+    /**
+     * Training algorithm for the dummy classifier. Actually dummy classifier returns the maximum occurring class in
+     * the training data, there is no training.
+     *
+     * @param trainSet   Training data given to the algorithm.
+     * @param parameters -
+     */
+    train(trainSet: InstanceList, parameters: Parameter): void;
+}

package/dist/Classifier/Dummy.js

@@ -0,0 +1,29 @@
+(function (factory) {
+    if (typeof module === "object" && typeof module.exports === "object") {
+        var v = factory(require, exports);
+        if (v !== undefined) module.exports = v;
+    }
+    else if (typeof define === "function" && define.amd) {
+        define(["require", "exports", "./Classifier", "../Model/DummyModel"], factory);
+    }
+})(function (require, exports) {
+    "use strict";
+    Object.defineProperty(exports, "__esModule", { value: true });
+    exports.Dummy = void 0;
+    const Classifier_1 = require("./Classifier");
+    const DummyModel_1 = require("../Model/DummyModel");
+    class Dummy extends Classifier_1.Classifier {
+        /**
+         * Training algorithm for the dummy classifier. Actually dummy classifier returns the maximum occurring class in
+         * the training data, there is no training.
+         *
+         * @param trainSet   Training data given to the algorithm.
+         * @param parameters -
+         */
+        train(trainSet, parameters) {
+            this.model = new DummyModel_1.DummyModel(trainSet);
+        }
+    }
+    exports.Dummy = Dummy;
+});
+//# sourceMappingURL=Dummy.js.map

package/dist/Classifier/Dummy.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"Dummy.js","sourceRoot":"","sources":["../../source/Classifier/Dummy.ts"],"names":[],"mappings":";;;;;;;;;;;;IAAA,6CAAwC;IAGxC,oDAA+C;IAE/C,MAAa,KAAM,SAAQ,uBAAU;QAEjC;;;;;;WAMG;QACH,KAAK,CAAC,QAAsB,EAAE,UAAqB;YAC/C,IAAI,CAAC,KAAK,GAAG,IAAI,uBAAU,CAAC,QAAQ,CAAC,CAAC;QAC1C,CAAC;KAEJ;IAbD,sBAaC"}

package/dist/Classifier/KMeans.d.ts

@@ -0,0 +1,12 @@
+import { Classifier } from "./Classifier";
+import { InstanceList } from "../InstanceList/InstanceList";
+import { Parameter } from "../Parameter/Parameter";
+export declare class KMeans extends Classifier {
+    /**
+     * Training algorithm for K-Means classifier. K-Means finds the mean of each class for training.
+     *
+     * @param trainSet   Training data given to the algorithm.
+     * @param parameters distanceMetric: distance metric used to calculate the distance between two instances.
+     */
+    train(trainSet: InstanceList, parameters: Parameter): void;
+}

package/dist/Classifier/KMeans.js

@@ -0,0 +1,36 @@
+(function (factory) {
+    if (typeof module === "object" && typeof module.exports === "object") {
+        var v = factory(require, exports);
+        if (v !== undefined) module.exports = v;
+    }
+    else if (typeof define === "function" && define.amd) {
+        define(["require", "exports", "./Classifier", "../InstanceList/InstanceList", "../InstanceList/Partition", "../Model/KMeansModel"], factory);
+    }
+})(function (require, exports) {
+    "use strict";
+    Object.defineProperty(exports, "__esModule", { value: true });
+    exports.KMeans = void 0;
+    const Classifier_1 = require("./Classifier");
+    const InstanceList_1 = require("../InstanceList/InstanceList");
+    const Partition_1 = require("../InstanceList/Partition");
+    const KMeansModel_1 = require("../Model/KMeansModel");
+    class KMeans extends Classifier_1.Classifier {
+        /**
+         * Training algorithm for K-Means classifier. K-Means finds the mean of each class for training.
+         *
+         * @param trainSet   Training data given to the algorithm.
+         * @param parameters distanceMetric: distance metric used to calculate the distance between two instances.
+         */
+        train(trainSet, parameters) {
+            let priorDistribution = trainSet.classDistribution();
+            let classMeans = new InstanceList_1.InstanceList();
+            let classLists = new Partition_1.Partition(trainSet);
+            for (let i = 0; i < classLists.size(); i++) {
+                classMeans.add(classLists.get(i).average());
+            }
+            this.model = new KMeansModel_1.KMeansModel(priorDistribution, classMeans, parameters.getDistanceMetric());
+        }
+    }
+    exports.KMeans = KMeans;
+});
+//# sourceMappingURL=KMeans.js.map

package/dist/Classifier/KMeans.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"KMeans.js","sourceRoot":"","sources":["../../source/Classifier/KMeans.ts"],"names":[],"mappings":";;;;;;;;;;;;IAAA,6CAAwC;IACxC,+DAA0D;IAE1D,yDAAoD;IACpD,sDAAiD;IAGjD,MAAa,MAAO,SAAQ,uBAAU;QAElC;;;;;WAKG;QACH,KAAK,CAAC,QAAsB,EAAE,UAAqB;YAC/C,IAAI,iBAAiB,GAAG,QAAQ,CAAC,iBAAiB,EAAE,CAAC;YACrD,IAAI,UAAU,GAAG,IAAI,2BAAY,EAAE,CAAC;YACpC,IAAI,UAAU,GAAG,IAAI,qBAAS,CAAC,QAAQ,CAAC,CAAC;YACzC,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,CAAC,IAAI,EAAE,EAAE,CAAC,EAAE,EAAE;gBACxC,UAAU,CAAC,GAAG,CAAC,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,EAAE,CAAC,CAAC;aAC/C;YACD,IAAI,CAAC,KAAK,GAAG,IAAI,yBAAW,CAAC,iBAAiB,EAAE,UAAU,EAAqB,UAAW,CAAC,iBAAiB,EAAE,CAAC,CAAC;QACpH,CAAC;KAEJ;IAlBD,wBAkBC"}

package/dist/Classifier/Knn.d.ts

@@ -0,0 +1,13 @@
+import { Classifier } from "./Classifier";
+import { InstanceList } from "../InstanceList/InstanceList";
+import { Parameter } from "../Parameter/Parameter";
+export declare class Knn extends Classifier {
+    /**
+     * Training algorithm for K-nearest neighbor classifier.
+     *
+     * @param trainSet   Training data given to the algorithm.
+     * @param parameters K: k parameter of the K-nearest neighbor algorithm
+     *                   distanceMetric: distance metric used to calculate the distance between two instances.
+     */
+    train(trainSet: InstanceList, parameters: Parameter): void;
+}

package/dist/Classifier/Knn.js

@@ -0,0 +1,29 @@
+(function (factory) {
+    if (typeof module === "object" && typeof module.exports === "object") {
+        var v = factory(require, exports);
+        if (v !== undefined) module.exports = v;
+    }
+    else if (typeof define === "function" && define.amd) {
+        define(["require", "exports", "./Classifier", "../Model/KnnModel"], factory);
+    }
+})(function (require, exports) {
+    "use strict";
+    Object.defineProperty(exports, "__esModule", { value: true });
+    exports.Knn = void 0;
+    const Classifier_1 = require("./Classifier");
+    const KnnModel_1 = require("../Model/KnnModel");
+    class Knn extends Classifier_1.Classifier {
+        /**
+         * Training algorithm for K-nearest neighbor classifier.
+         *
+         * @param trainSet   Training data given to the algorithm.
+         * @param parameters K: k parameter of the K-nearest neighbor algorithm
+         *                   distanceMetric: distance metric used to calculate the distance between two instances.
+         */
+        train(trainSet, parameters) {
+            this.model = new KnnModel_1.KnnModel(trainSet, parameters.getK(), parameters.getDistanceMetric());
+        }
+    }
+    exports.Knn = Knn;
+});
+//# sourceMappingURL=Knn.js.map

package/dist/Classifier/Knn.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"Knn.js","sourceRoot":"","sources":["../../source/Classifier/Knn.ts"],"names":[],"mappings":";;;;;;;;;;;;IAAA,6CAAwC;IAGxC,gDAA2C;IAG3C,MAAa,GAAI,SAAQ,uBAAU;QAG/B;;;;;;WAMG;QACH,KAAK,CAAC,QAAsB,EAAE,UAAqB;YAC/C,IAAI,CAAC,KAAK,GAAG,IAAI,mBAAQ,CAAC,QAAQ,EAAkB,UAAW,CAAC,IAAI,EAAE,EAAkB,UAAW,CAAC,iBAAiB,EAAE,CAAC,CAAC;QAC7H,CAAC;KAEJ;IAdD,kBAcC"}

package/dist/Classifier/Lda.d.ts

@@ -0,0 +1,12 @@
+import { Classifier } from "./Classifier";
+import { InstanceList } from "../InstanceList/InstanceList";
+import { Parameter } from "../Parameter/Parameter";
+export declare class Lda extends Classifier {
+    /**
+     * Training algorithm for the linear discriminant analysis classifier (Introduction to Machine Learning, Alpaydin, 2015).
+     *
+     * @param trainSet   Training data given to the algorithm.
+     * @param parameters -
+     */
+    train(trainSet: InstanceList, parameters: Parameter): void;
+}

package/dist/Classifier/Lda.js

@@ -0,0 +1,52 @@
+(function (factory) {
+    if (typeof module === "object" && typeof module.exports === "object") {
+        var v = factory(require, exports);
+        if (v !== undefined) module.exports = v;
+    }
+    else if (typeof define === "function" && define.amd) {
+        define(["require", "exports", "./Classifier", "nlptoolkit-math/dist/Vector", "../InstanceList/Partition", "nlptoolkit-math/dist/Matrix", "../Model/LdaModel"], factory);
+    }
+})(function (require, exports) {
+    "use strict";
+    Object.defineProperty(exports, "__esModule", { value: true });
+    exports.Lda = void 0;
+    const Classifier_1 = require("./Classifier");
+    const Vector_1 = require("nlptoolkit-math/dist/Vector");
+    const Partition_1 = require("../InstanceList/Partition");
+    const Matrix_1 = require("nlptoolkit-math/dist/Matrix");
+    const LdaModel_1 = require("../Model/LdaModel");
+    class Lda extends Classifier_1.Classifier {
+        /**
+         * Training algorithm for the linear discriminant analysis classifier (Introduction to Machine Learning, Alpaydin, 2015).
+         *
+         * @param trainSet   Training data given to the algorithm.
+         * @param parameters -
+         */
+        train(trainSet, parameters) {
+            let w0 = new Map();
+            let w = new Map();
+            let priorDistribution = trainSet.classDistribution();
+            let classLists = new Partition_1.Partition(trainSet);
+            let covariance = new Matrix_1.Matrix(trainSet.get(0).continuousAttributeSize(), trainSet.get(0).continuousAttributeSize());
+            for (let i = 0; i < classLists.size(); i++) {
+                let averageVector = new Vector_1.Vector(classLists.get(i).continuousAttributeAverage());
+                let classCovariance = classLists.get(i).covariance(averageVector);
+                classCovariance.multiplyWithConstant(classLists.get(i).size() - 1);
+                covariance.add(classCovariance);
+            }
+            covariance.divideByConstant(trainSet.size() - classLists.size());
+            covariance.inverse();
+            for (let i = 0; i < classLists.size(); i++) {
+                let Ci = classLists.get(i).getClassLabel();
+                let averageVector = new Vector_1.Vector(classLists.get(i).continuousAttributeAverage());
+                let wi = covariance.multiplyWithVectorFromRight(averageVector);
+                w.set(Ci, wi);
+                let w0i = -0.5 * wi.dotProduct(averageVector) + Math.log(priorDistribution.getProbability(Ci));
+                w0.set(Ci, w0i);
+            }
+            this.model = new LdaModel_1.LdaModel(priorDistribution, w, w0);
+        }
+    }
+    exports.Lda = Lda;
+});
+//# sourceMappingURL=Lda.js.map

package/dist/Classifier/Lda.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"Lda.js","sourceRoot":"","sources":["../../source/Classifier/Lda.ts"],"names":[],"mappings":";;;;;;;;;;;;IAAA,6CAAwC;IAGxC,wDAAmD;IACnD,yDAAoD;IACpD,wDAAmD;IAEnD,gDAA2C;IAE3C,MAAa,GAAI,SAAQ,uBAAU;QAE/B;;;;;WAKG;QACH,KAAK,CAAC,QAAsB,EAAE,UAAqB;YAC/C,IAAI,EAAE,GAAG,IAAI,GAAG,EAAkB,CAAC;YACnC,IAAI,CAAC,GAAG,IAAI,GAAG,EAAkB,CAAC;YAClC,IAAI,iBAAiB,GAAG,QAAQ,CAAC,iBAAiB,EAAE,CAAC;YACrD,IAAI,UAAU,GAAG,IAAI,qBAAS,CAAC,QAAQ,CAAC,CAAC;YACzC,IAAI,UAAU,GAAG,IAAI,eAAM,CAAC,QAAQ,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,uBAAuB,EAAE,EAAE,QAAQ,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,uBAAuB,EAAE,CAAC,CAAC;YAClH,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,CAAC,IAAI,EAAE,EAAE,CAAC,EAAE,EAAE;gBACxC,IAAI,aAAa,GAAG,IAAI,eAAM,CAAC,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,0BAA0B,EAAE,CAAC,CAAC;gBAC/E,IAAI,eAAe,GAAG,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,UAAU,CAAC,aAAa,CAAC,CAAC;gBAClE,eAAe,CAAC,oBAAoB,CAAC,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,IAAI,EAAE,GAAG,CAAC,CAAC,CAAC;gBACnE,UAAU,CAAC,GAAG,CAAC,eAAe,CAAC,CAAC;aACnC;YACD,UAAU,CAAC,gBAAgB,CAAC,QAAQ,CAAC,IAAI,EAAE,GAAG,UAAU,CAAC,IAAI,EAAE,CAAC,CAAC;YACjE,UAAU,CAAC,OAAO,EAAE,CAAC;YACrB,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,CAAC,IAAI,EAAE,EAAE,CAAC,EAAE,EAAE;gBACxC,IAAI,EAAE,GAA8B,UAAU,CAAC,GAAG,CAAC,CAAC,CAAE,CAAC,aAAa,EAAE,CAAC;gBACvE,IAAI,aAAa,GAAG,IAAI,eAAM,CAAC,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,0BAA0B,EAAE,CAAC,CAAC;gBAC/E,IAAI,EAAE,GAAG,UAAU,CAAC,2BAA2B,CAAC,aAAa,CAAC,CAAC;gBAC/D,CAAC,CAAC,GAAG,CAAC,EAAE,EAAE,EAAE,CAAC,CAAC;gBACd,IAAI,GAAG,GAAG,CAAC,GAAG,GAAG,EAAE,CAAC,UAAU,CAAC,aAAa,CAAC,GAAG,IAAI,CAAC,GAAG,CAAC,iBAAiB,CAAC,cAAc,CAAC,EAAE,CAAC,CAAC,CAAC;gBAC/F,EAAE,CAAC,GAAG,CAAC,EAAE,EAAE,GAAG,CAAC,CAAC;aACnB;YACD,IAAI,CAAC,KAAK,GAAG,IAAI,mBAAQ,CAAC,iBAAiB,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC;QACxD,CAAC;KAEJ;IAjCD,kBAiCC"}

package/dist/Classifier/LinearPerceptron.d.ts

@@ -0,0 +1,14 @@
+import { Classifier } from "./Classifier";
+import { InstanceList } from "../InstanceList/InstanceList";
+import { Parameter } from "../Parameter/Parameter";
+export declare class LinearPerceptron extends Classifier {
+    /**
+     * Training algorithm for the linear perceptron algorithm. 20 percent of the data is separated as cross-validation
+     * data used for selecting the best weights. 80 percent of the data is used for training the linear perceptron with
+     * gradient descent.
+     *
+     * @param trainSet   Training data given to the algorithm
+     * @param parameters Parameters of the linear perceptron.
+     */
+    train(trainSet: InstanceList, parameters: Parameter): void;
+}

package/dist/Classifier/LinearPerceptron.js

@@ -0,0 +1,32 @@
+(function (factory) {
+    if (typeof module === "object" && typeof module.exports === "object") {
+        var v = factory(require, exports);
+        if (v !== undefined) module.exports = v;
+    }
+    else if (typeof define === "function" && define.amd) {
+        define(["require", "exports", "./Classifier", "../InstanceList/Partition", "../Model/LinearPerceptronModel"], factory);
+    }
+})(function (require, exports) {
+    "use strict";
+    Object.defineProperty(exports, "__esModule", { value: true });
+    exports.LinearPerceptron = void 0;
+    const Classifier_1 = require("./Classifier");
+    const Partition_1 = require("../InstanceList/Partition");
+    const LinearPerceptronModel_1 = require("../Model/LinearPerceptronModel");
+    class LinearPerceptron extends Classifier_1.Classifier {
+        /**
+         * Training algorithm for the linear perceptron algorithm. 20 percent of the data is separated as cross-validation
+         * data used for selecting the best weights. 80 percent of the data is used for training the linear perceptron with
+         * gradient descent.
+         *
+         * @param trainSet   Training data given to the algorithm
+         * @param parameters Parameters of the linear perceptron.
+         */
+        train(trainSet, parameters) {
+            let partition = new Partition_1.Partition(trainSet, parameters.getCrossValidationRatio(), true);
+            this.model = new LinearPerceptronModel_1.LinearPerceptronModel(partition.get(1), partition.get(0), parameters);
+        }
+    }
+    exports.LinearPerceptron = LinearPerceptron;
+});
+//# sourceMappingURL=LinearPerceptron.js.map

package/dist/Classifier/LinearPerceptron.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"LinearPerceptron.js","sourceRoot":"","sources":["../../source/Classifier/LinearPerceptron.ts"],"names":[],"mappings":";;;;;;;;;;;;IAAA,6CAAwC;IAGxC,yDAAoD;IAEpD,0EAAqE;IAErE,MAAa,gBAAiB,SAAQ,uBAAU;QAE5C;;;;;;;WAOG;QACH,KAAK,CAAC,QAAsB,EAAE,UAAqB;YAC/C,IAAI,SAAS,GAAG,IAAI,qBAAS,CAAC,QAAQ,EAA+B,UAAW,CAAC,uBAAuB,EAAE,EAAE,IAAI,CAAC,CAAC;YAClH,IAAI,CAAC,KAAK,GAAG,IAAI,6CAAqB,CAAC,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,EAA8B,UAAU,CAAC,CAAC;QACvH,CAAC;KAEJ;IAfD,4CAeC"}

package/dist/Classifier/MultiLayerPerceptron.d.ts

@@ -0,0 +1,14 @@
+import { Classifier } from "./Classifier";
+import { InstanceList } from "../InstanceList/InstanceList";
+import { Parameter } from "../Parameter/Parameter";
+export declare class MultiLayerPerceptron extends Classifier {
+    /**
+     * Training algorithm for the multilayer perceptron algorithm. 20 percent of the data is separated as cross-validation
+     * data used for selecting the best weights. 80 percent of the data is used for training the multilayer perceptron with
+     * gradient descent.
+     *
+     * @param trainSet   Training data given to the algorithm
+     * @param parameters Parameters of the multilayer perceptron.
+     */
+    train(trainSet: InstanceList, parameters: Parameter): void;
+}

package/dist/Classifier/MultiLayerPerceptron.js

@@ -0,0 +1,32 @@
+(function (factory) {
+    if (typeof module === "object" && typeof module.exports === "object") {
+        var v = factory(require, exports);
+        if (v !== undefined) module.exports = v;
+    }
+    else if (typeof define === "function" && define.amd) {
+        define(["require", "exports", "./Classifier", "../InstanceList/Partition", "../Model/MultiLayerPerceptronModel"], factory);
+    }
+})(function (require, exports) {
+    "use strict";
+    Object.defineProperty(exports, "__esModule", { value: true });
+    exports.MultiLayerPerceptron = void 0;
+    const Classifier_1 = require("./Classifier");
+    const Partition_1 = require("../InstanceList/Partition");
+    const MultiLayerPerceptronModel_1 = require("../Model/MultiLayerPerceptronModel");
+    class MultiLayerPerceptron extends Classifier_1.Classifier {
+        /**
+         * Training algorithm for the multilayer perceptron algorithm. 20 percent of the data is separated as cross-validation
+         * data used for selecting the best weights. 80 percent of the data is used for training the multilayer perceptron with
+         * gradient descent.
+         *
+         * @param trainSet   Training data given to the algorithm
+         * @param parameters Parameters of the multilayer perceptron.
+         */
+        train(trainSet, parameters) {
+            let partition = new Partition_1.Partition(trainSet, parameters.getCrossValidationRatio(), true);
+            this.model = new MultiLayerPerceptronModel_1.MultiLayerPerceptronModel(partition.get(1), partition.get(0), parameters);
+        }
+    }
+    exports.MultiLayerPerceptron = MultiLayerPerceptron;
+});
+//# sourceMappingURL=MultiLayerPerceptron.js.map

package/dist/Classifier/MultiLayerPerceptron.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"MultiLayerPerceptron.js","sourceRoot":"","sources":["../../source/Classifier/MultiLayerPerceptron.ts"],"names":[],"mappings":";;;;;;;;;;;;IAAA,6CAAwC;IAGxC,yDAAoD;IAEpD,kFAA6E;IAE7E,MAAa,oBAAqB,SAAQ,uBAAU;QAEhD;;;;;;;WAOG;QACH,KAAK,CAAC,QAAsB,EAAE,UAAqB;YAC/C,IAAI,SAAS,GAAG,IAAI,qBAAS,CAAC,QAAQ,EAAmC,UAAW,CAAC,uBAAuB,EAAE,EAAE,IAAI,CAAC,CAAC;YACtH,IAAI,CAAC,KAAK,GAAG,IAAI,qDAAyB,CAAC,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,EAAE,SAAS,CAAC,GAAG,CAAC,CAAC,CAAC,EAAkC,UAAU,CAAC,CAAC;QAC/H,CAAC;KAEJ;IAfD,oDAeC"}

package/dist/Classifier/NaiveBayes.d.ts

@@ -0,0 +1,25 @@
+import { Classifier } from "./Classifier";
+import { InstanceList } from "../InstanceList/InstanceList";
+import { Parameter } from "../Parameter/Parameter";
+export declare class NaiveBayes extends Classifier {
+    /**
+     * Training algorithm for Naive Bayes algorithm with a continuous data set.
+     *
+     * @param priorDistribution Probability distribution of classes P(C_i)
+     * @param classLists        Instances are divided into K lists, where each list contains only instances from a single class
+     */
+    private trainContinuousVersion;
+    /**
+     * Training algorithm for Naive Bayes algorithm with a discrete data set.
+     * @param priorDistribution Probability distribution of classes P(C_i)
+     * @param classLists Instances are divided into K lists, where each list contains only instances from a single class
+     */
+    private trainDiscreteVersion;
+    /**
+     * Training algorithm for Naive Bayes algorithm. It basically calls trainContinuousVersion for continuous data sets,
+     * trainDiscreteVersion for discrete data sets.
+     * @param trainSet Training data given to the algorithm
+     * @param parameters -
+     */
+    train(trainSet: InstanceList, parameters: Parameter): void;
+}

package/dist/Classifier/NaiveBayes.js

@@ -0,0 +1,67 @@
+(function (factory) {
+    if (typeof module === "object" && typeof module.exports === "object") {
+        var v = factory(require, exports);
+        if (v !== undefined) module.exports = v;
+    }
+    else if (typeof define === "function" && define.amd) {
+        define(["require", "exports", "./Classifier", "../InstanceList/Partition", "../Model/NaiveBayesModel", "../Attribute/DiscreteAttribute"], factory);
+    }
+})(function (require, exports) {
+    "use strict";
+    Object.defineProperty(exports, "__esModule", { value: true });
+    exports.NaiveBayes = void 0;
+    const Classifier_1 = require("./Classifier");
+    const Partition_1 = require("../InstanceList/Partition");
+    const NaiveBayesModel_1 = require("../Model/NaiveBayesModel");
+    const DiscreteAttribute_1 = require("../Attribute/DiscreteAttribute");
+    class NaiveBayes extends Classifier_1.Classifier {
+        /**
+         * Training algorithm for Naive Bayes algorithm with a continuous data set.
+         *
+         * @param priorDistribution Probability distribution of classes P(C_i)
+         * @param classLists        Instances are divided into K lists, where each list contains only instances from a single class
+         */
+        trainContinuousVersion(priorDistribution, classLists) {
+            let classMeans = new Map();
+            let classDeviations = new Map();
+            for (let i = 0; i < classLists.size(); i++) {
+                let classLabel = classLists.get(i).getClassLabel();
+                let averageVector = classLists.get(i).average().toVector();
+                classMeans.set(classLabel, averageVector);
+                let standardDeviationVector = classLists.get(i).standardDeviation().toVector();
+                classDeviations.set(classLabel, standardDeviationVector);
+            }
+            this.model = new NaiveBayesModel_1.NaiveBayesModel(priorDistribution, classMeans, classDeviations);
+        }
+        /**
+         * Training algorithm for Naive Bayes algorithm with a discrete data set.
+         * @param priorDistribution Probability distribution of classes P(C_i)
+         * @param classLists Instances are divided into K lists, where each list contains only instances from a single class
+         */
+        trainDiscreteVersion(priorDistribution, classLists) {
+            let classAttributeDistributions = new Map();
+            for (let i = 0; i < classLists.size(); i++) {
+                classAttributeDistributions.set(classLists.get(i).getClassLabel(), classLists.get(i).allAttributesDistribution());
+            }
+            this.model = new NaiveBayesModel_1.NaiveBayesModel(priorDistribution, classAttributeDistributions);
+        }
+        /**
+         * Training algorithm for Naive Bayes algorithm. It basically calls trainContinuousVersion for continuous data sets,
+         * trainDiscreteVersion for discrete data sets.
+         * @param trainSet Training data given to the algorithm
+         * @param parameters -
+         */
+        train(trainSet, parameters) {
+            let priorDistribution = trainSet.classDistribution();
+            let classLists = new Partition_1.Partition(trainSet);
+            if (classLists.get(0).get(0).getAttribute(0) instanceof DiscreteAttribute_1.DiscreteAttribute) {
+                this.trainDiscreteVersion(priorDistribution, classLists);
+            }
+            else {
+                this.trainContinuousVersion(priorDistribution, classLists);
+            }
+        }
+    }
+    exports.NaiveBayes = NaiveBayes;
+});
+//# sourceMappingURL=NaiveBayes.js.map

package/dist/Classifier/NaiveBayes.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"NaiveBayes.js","sourceRoot":"","sources":["../../source/Classifier/NaiveBayes.ts"],"names":[],"mappings":";;;;;;;;;;;;IAAA,6CAAwC;IAIxC,yDAAoD;IAGpD,8DAAyD;IACzD,sEAAiE;IAEjE,MAAa,UAAW,SAAQ,uBAAU;QAEtC;;;;;WAKG;QACK,sBAAsB,CAAC,iBAAuC,EAAE,UAAqB;YACzF,IAAI,UAAU,GAAG,IAAI,GAAG,EAAkB,CAAC;YAC3C,IAAI,eAAe,GAAG,IAAI,GAAG,EAAkB,CAAC;YAChD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,CAAC,IAAI,EAAE,EAAE,CAAC,EAAE,EAAC;gBACvC,IAAI,UAAU,GAA8B,UAAU,CAAC,GAAG,CAAC,CAAC,CAAE,CAAC,aAAa,EAAE,CAAC;gBAC/E,IAAI,aAAa,GAAG,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,OAAO,EAAE,CAAC,QAAQ,EAAE,CAAC;gBAC3D,UAAU,CAAC,GAAG,CAAC,UAAU,EAAE,aAAa,CAAC,CAAC;gBAC1C,IAAI,uBAAuB,GAAG,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,iBAAiB,EAAE,CAAC,QAAQ,EAAE,CAAC;gBAC/E,eAAe,CAAC,GAAG,CAAC,UAAU,EAAE,uBAAuB,CAAC,CAAC;aAC5D;YACD,IAAI,CAAC,KAAK,GAAG,IAAI,iCAAe,CAAC,iBAAiB,EAAE,UAAU,EAAE,eAAe,CAAC,CAAC;QACrF,CAAC;QAED;;;;WAIG;QACK,oBAAoB,CAAC,iBAAuC,EAAE,UAAqB;YACvF,IAAI,2BAA2B,GAAG,IAAI,GAAG,EAAuC,CAAC;YACjF,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,CAAC,IAAI,EAAE,EAAE,CAAC,EAAE,EAAC;gBACvC,2BAA2B,CAAC,GAAG,CAA4B,UAAU,CAAC,GAAG,CAAC,CAAC,CAAE,CAAC,aAAa,EAAE,EACzF,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,yBAAyB,EAAE,CAAC,CAAC;aACtD;YACD,IAAI,CAAC,KAAK,GAAG,IAAI,iCAAe,CAAC,iBAAiB,EAAE,2BAA2B,CAAC,CAAC;QACrF,CAAC;QAED;;;;;WAKG;QACH,KAAK,CAAC,QAAsB,EAAE,UAAqB;YAC/C,IAAI,iBAAiB,GAAG,QAAQ,CAAC,iBAAiB,EAAE,CAAC;YACrD,IAAI,UAAU,GAAG,IAAI,qBAAS,CAAC,QAAQ,CAAC,CAAC;YACzC,IAAI,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,YAAY,CAAC,CAAC,CAAC,YAAY,qCAAiB,EAAC;gBACtE,IAAI,CAAC,oBAAoB,CAAC,iBAAiB,EAAE,UAAU,CAAC,CAAC;aAC5D;iBAAM;gBACH,IAAI,CAAC,sBAAsB,CAAC,iBAAiB,EAAE,UAAU,CAAC,CAAC;aAC9D;QACL,CAAC;KAEJ;IAnDD,gCAmDC"}

package/dist/Classifier/Qda.d.ts

@@ -0,0 +1,12 @@
+import { Classifier } from "./Classifier";
+import { Parameter } from "../Parameter/Parameter";
+import { InstanceList } from "../InstanceList/InstanceList";
+export declare class Qda extends Classifier {
+    /**
+     * Training algorithm for the quadratic discriminant analysis classifier (Introduction to Machine Learning, Alpaydin, 2015).
+     *
+     * @param trainSet   Training data given to the algorithm.
+     * @param parameters -
+     */
+    train(trainSet: InstanceList, parameters: Parameter): void;
+}

package/dist/Classifier/Qda.js

@@ -0,0 +1,50 @@
+(function (factory) {
+    if (typeof module === "object" && typeof module.exports === "object") {
+        var v = factory(require, exports);
+        if (v !== undefined) module.exports = v;
+    }
+    else if (typeof define === "function" && define.amd) {
+        define(["require", "exports", "./Classifier", "nlptoolkit-math/dist/Vector", "../InstanceList/Partition", "../Model/QdaModel"], factory);
+    }
+})(function (require, exports) {
+    "use strict";
+    Object.defineProperty(exports, "__esModule", { value: true });
+    exports.Qda = void 0;
+    const Classifier_1 = require("./Classifier");
+    const Vector_1 = require("nlptoolkit-math/dist/Vector");
+    const Partition_1 = require("../InstanceList/Partition");
+    const QdaModel_1 = require("../Model/QdaModel");
+    class Qda extends Classifier_1.Classifier {
+        /**
+         * Training algorithm for the quadratic discriminant analysis classifier (Introduction to Machine Learning, Alpaydin, 2015).
+         *
+         * @param trainSet   Training data given to the algorithm.
+         * @param parameters -
+         */
+        train(trainSet, parameters) {
+            let determinant = 0;
+            let w0 = new Map();
+            let w = new Map();
+            let W = new Map();
+            let classLists = new Partition_1.Partition(trainSet);
+            let priorDistribution = trainSet.classDistribution();
+            for (let i = 0; i < classLists.size(); i++) {
+                let Ci = classLists.get(i).getClassLabel();
+                let averageVector = new Vector_1.Vector(classLists.get(i).continuousAttributeAverage());
+                let classCovariance = classLists.get(i).covariance(averageVector);
+                determinant = classCovariance.determinant();
+                classCovariance.inverse();
+                let Wi = classCovariance.clone();
+                Wi.multiplyWithConstant(-0.5);
+                W.set(Ci, Wi);
+                let wi = classCovariance.multiplyWithVectorFromLeft(averageVector);
+                w.set(Ci, wi);
+                let w0i = -0.5 * (wi.dotProduct(averageVector) + Math.log(determinant)) + Math.log(priorDistribution.getProbability(Ci));
+                w0.set(Ci, w0i);
+            }
+            this.model = new QdaModel_1.QdaModel(priorDistribution, W, w, w0);
+        }
+    }
+    exports.Qda = Qda;
+});
+//# sourceMappingURL=Qda.js.map

package/dist/Classifier/Qda.js.map

@@ -0,0 +1 @@
+{"version":3,"file":"Qda.js","sourceRoot":"","sources":["../../source/Classifier/Qda.ts"],"names":[],"mappings":";;;;;;;;;;;;IAAA,6CAAwC;IAGxC,wDAAmD;IAEnD,yDAAoD;IAEpD,gDAA2C;IAE3C,MAAa,GAAI,SAAQ,uBAAU;QAE/B;;;;;WAKG;QACH,KAAK,CAAC,QAAsB,EAAE,UAAqB;YAC/C,IAAI,WAAW,GAAG,CAAC,CAAA;YACnB,IAAI,EAAE,GAAG,IAAI,GAAG,EAAkB,CAAC;YACnC,IAAI,CAAC,GAAG,IAAI,GAAG,EAAkB,CAAC;YAClC,IAAI,CAAC,GAAG,IAAI,GAAG,EAAkB,CAAC;YAClC,IAAI,UAAU,GAAG,IAAI,qBAAS,CAAC,QAAQ,CAAC,CAAC;YACzC,IAAI,iBAAiB,GAAG,QAAQ,CAAC,iBAAiB,EAAE,CAAC;YACrD,KAAK,IAAI,CAAC,GAAG,CAAC,EAAE,CAAC,GAAG,UAAU,CAAC,IAAI,EAAE,EAAE,CAAC,EAAE,EAAE;gBACxC,IAAI,EAAE,GAA8B,UAAU,CAAC,GAAG,CAAC,CAAC,CAAE,CAAC,aAAa,EAAE,CAAC;gBACvE,IAAI,aAAa,GAAG,IAAI,eAAM,CAAC,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,0BAA0B,EAAE,CAAC,CAAC;gBAC/E,IAAI,eAAe,GAAG,UAAU,CAAC,GAAG,CAAC,CAAC,CAAC,CAAC,UAAU,CAAC,aAAa,CAAC,CAAC;gBAClE,WAAW,GAAG,eAAe,CAAC,WAAW,EAAE,CAAC;gBAC5C,eAAe,CAAC,OAAO,EAAE,CAAC;gBAC1B,IAAI,EAAE,GAAG,eAAe,CAAC,KAAK,EAAE,CAAC;gBACjC,EAAE,CAAC,oBAAoB,CAAC,CAAC,GAAG,CAAC,CAAC;gBAC9B,CAAC,CAAC,GAAG,CAAC,EAAE,EAAE,EAAE,CAAC,CAAC;gBACd,IAAI,EAAE,GAAG,eAAe,CAAC,0BAA0B,CAAC,aAAa,CAAC,CAAC;gBACnE,CAAC,CAAC,GAAG,CAAC,EAAE,EAAE,EAAE,CAAC,CAAC;gBACd,IAAI,GAAG,GAAG,CAAC,GAAG,GAAG,CAAC,EAAE,CAAC,UAAU,CAAC,aAAa,CAAC,GAAG,IAAI,CAAC,GAAG,CAAC,WAAW,CAAC,CAAC,GAAG,IAAI,CAAC,GAAG,CAAC,iBAAiB,CAAC,cAAc,CAAC,EAAE,CAAC,CAAC,CAAC;gBACzH,EAAE,CAAC,GAAG,CAAC,EAAE,EAAE,GAAG,CAAC,CAAC;aACnB;YACD,IAAI,CAAC,KAAK,GAAG,IAAI,mBAAQ,CAAC,iBAAiB,EAAE,CAAC,EAAE,CAAC,EAAE,EAAE,CAAC,CAAC;QAC3D,CAAC;KAEJ;IAhCD,kBAgCC"}