Add particle.js

Author: Jasonnor

Particle-Swarm-Optimization/js/particle.js

@@ -0,0 +1,70 @@
+var Particle = {
+    // xLength: 1 + numberOfNeurons + numberOfNeurons * dimension + numberOfNeurons
+    createNew: function () {
+        var particle = {};
+        particle.x = [];
+        particle.v = [];
+        particle.fitness = 10000;
+        particle.rbf = RBF.createNew();
+        particle.clone = function () {
+            var particleNew = Particle.createNew();
+            for (var i = 0; i < this.x.length; i++)
+                particleNew.x[i] = this.x[i];
+            for (var i = 0; i < this.v.length; i++)
+                particleNew.v[i] = this.v[i];
+            particleNew.normalization();
+            return particleNew;
+        };
+        particle.move = function () {
+            for (var i = 0; i < this.x.length; i++)
+                x[i] += v[i];
+        };
+        particle.normalization = function (particle) {
+            this.x[0] = Math.min(Math.max(this.x[0], 0), 1);
+            this.rbf.theta = this.x[0];
+            for (var i = 0; i < numberOfNeurons; i++) {
+                this.x[i + 1] = Math.min(Math.max(this.x[i + 1], -40), 40);
+                this.rbf.W[i] = this.x[i + 1];
+            }
+            for (var i = 1 + numberOfNeurons, j = 0; i < 1 + numberOfNeurons + numberOfNeurons * dimension; i++, j++) {
+                this.x[i] = Math.min(Math.max(this.x[i], 0), 30);
+                this.rbf.M[parseInt(j / dimension)][parseInt(j % dimension)] = this.x[i];
+            }
+            for (var i = 1 + numberOfNeurons + numberOfNeurons * dimension, j = 0; i < 1 + numberOfNeurons + numberOfNeurons * dimension + numberOfNeurons; i++, j++) {
+                this.x[i] = Math.min(Math.max(this.x[i], 0.000001), 10);
+                this.rbf.sigma[j] = this.x[i];
+            }
+        };
+        particle.randomBuild = function () {
+            this.x[0] = Math.random();
+            for (var i = 1; i < 1 + numberOfNeurons; i++) {
+                this.x[i] = Math.random();
+            }
+            for (var i = 1 + numberOfNeurons; i < 1 + numberOfNeurons + numberOfNeurons * dimension; i++) {
+                this.x[i] = Math.random() * 30;
+            }
+            for (var i = 1 + numberOfNeurons + numberOfNeurons * dimension; i < 1 + numberOfNeurons + numberOfNeurons * dimension + numberOfNeurons; i++) {
+                this.x[i] = Math.random() * 10;
+            }
+            this.normalization();
+        };
+        particle.getFitness = function (yD, x) {
+            this.normalization();
+            var value = 0;
+            for (var i = 0; i < yD.length; i++) {
+                var fX = this.rbf.getOutput(x[i]);
+                value += Math.pow(yD[i] - fX, 2);
+            }
+            value = value / 2.0;
+            this.fitness = value;
+            return value;
+        };
+        particle.distanceTo = function(target) {
+            var distance = 0;
+            for (var i = 0; i < this.x.length; i++)
+                distance += (this.x[i] - target.x[i]) * (this.x[i] - target.x[i]);
+            return distance;
+        }
+        return particle;
+    }
+};