From f735fa931d6501bbe52e357568ff33617862b401 Mon Sep 17 00:00:00 2001
From: Eric Lin <ericlin071906@gmail.com>
Date: Fri, 21 Oct 2022 18:02:39 -0400
Subject: [PATCH] create LinearRegression

---
 .../math/Regression/LinearRegression.java     | 42 +++++++++++++++++++
 .../stuylib/util/plot/Playground.java         | 38 +++++++++++------
 2 files changed, 68 insertions(+), 12 deletions(-)
 create mode 100644 src/com/stuypulse/stuylib/math/Regression/LinearRegression.java

diff --git a/src/com/stuypulse/stuylib/math/Regression/LinearRegression.java b/src/com/stuypulse/stuylib/math/Regression/LinearRegression.java
new file mode 100644
index 00000000..b96a2c7d
--- /dev/null
+++ b/src/com/stuypulse/stuylib/math/Regression/LinearRegression.java
@@ -0,0 +1,42 @@
+package com.stuypulse.stuylib.math.Regression;
+
+import com.stuypulse.stuylib.math.Vector2D;
+
+// sum of x 
+public class LinearRegression {
+
+    public Vector2D[] points;
+
+    public LinearRegression(Vector2D... points){
+        this.points = points;
+    }
+
+    private Vector2D equation(){
+        double sumX = 0;
+        double sumY = 0;
+        double sumXX = 0;
+        double sumXY = 0;
+
+        for(Vector2D point : points) {
+            sumX += point.x;
+            sumY += point.y;
+            sumXX += Math.pow(point.x, 2);
+            sumXY += point.x * point.y;
+            
+        }
+
+        double intercept = (sumY * sumXX - sumX * sumXY) / (sumXX - Math.pow(sumX, 2));
+        double slope = (points.length * sumXY - sumX * sumY) / (points.length * sumXX - Math.pow(sumX,2));
+
+        return new Vector2D(slope, intercept);
+    }
+
+    public void addRefPoint(double x, double y){
+        
+    }
+
+    public double predictedValue(double x){
+        return equation().x * x + equation().y;
+    }
+
+}
diff --git a/src/com/stuypulse/stuylib/util/plot/Playground.java b/src/com/stuypulse/stuylib/util/plot/Playground.java
index 7e045404..b9e1c939 100644
--- a/src/com/stuypulse/stuylib/util/plot/Playground.java
+++ b/src/com/stuypulse/stuylib/util/plot/Playground.java
@@ -5,6 +5,7 @@
 package com.stuypulse.stuylib.util.plot;
 
 import com.stuypulse.stuylib.math.Vector2D;
+import com.stuypulse.stuylib.math.Regression.LinearRegression;
 import com.stuypulse.stuylib.math.interpolation.*;
 import com.stuypulse.stuylib.streams.*;
 import com.stuypulse.stuylib.streams.angles.AFuser;
@@ -34,11 +35,11 @@ public interface Constants {
         int WIDTH = 800;
         int HEIGHT = 600;
 
-        double MIN_X = -1.0;
-        double MAX_X = 1.0;
+        double MIN_X = 0.0;
+        double MAX_X = 10.0;
 
-        double MIN_Y = -1.0;
-        double MAX_Y = 1.0;
+        double MIN_Y = 0.0;
+        double MAX_Y = 10.0;
 
         Settings SETTINGS =
                 new Settings()
@@ -69,6 +70,11 @@ public static Series make(String id, BStream series) {
     }
 
     public static void main(String[] args) throws InterruptedException {
+
+
+
+
+
         Plot plot = new Plot(Constants.SETTINGS);
 
         VStream m = VStream.create(() -> plot.getMouse().sub(new Vector2D(0.5, 0.5)).mul(2));
@@ -95,14 +101,22 @@ public static void main(String[] args) throws InterruptedException {
         VStream delay = VStream.create(() -> delay_angle.get().getVector().mul(1.1));
         VStream afuser = VStream.create(() -> afuser_angle.get().getVector().mul(1.05));
 
-        plot.addSeries(Constants.make("Angle", mouse))
-                .addSeries(Constants.make("Jerk", jerk))
-                .addSeries(Constants.make("Rate", rate))
-                .addSeries(Constants.make("LPF", lpf))
-                .addSeries(Constants.make("HPF", hpf))
-                .addSeries(Constants.make("afuser", afuser))
-                .addSeries(Constants.make("delayed", delay))
-                .addSeries(Constants.make("mouse", m));
+        Vector2D[] refPoints = {new Vector2D(1,2), new  Vector2D(2,5), new Vector2D(6, 10)};
+        LinearRegression LinearRegression = new LinearRegression(refPoints);
+        plot.addSeries(new FuncSeries(
+            new Config("linear regression", 1000),
+            new Domain(0, 10),
+            x -> LinearRegression.predictedValue(x)
+        ));
+
+        // plot.addSeries(Constants.make("Angle", mouse))
+        //         .addSeries(Constants.make("Jerk", jerk))
+        //         .addSeries(Constants.make("Rate", rate))
+        //         .addSeries(Constants.make("LPF", lpf))
+        //         .addSeries(Constants.make("HPF", hpf))
+        //         .addSeries(Constants.make("afuser", afuser))
+        //         .addSeries(Constants.make("delayed", delay))
+        //         .addSeries(Constants.make("mouse", m));
         //
         // .addSeries(Constants.make("y=x", x -> x))
         // .addSeries(