Merge pull request #65 from jeffin07/master

Simple Linear Regression

Author: master-fury

Machine_Learning/SimpleLinearRegression.py

@@ -0,0 +1,33 @@
+import numpy as np
+import matplotlib.pyplot as plt  
+class SimpleLinearRegression:
+	def __init__(self):
+		self.w_0 = 0
+		self.w_1 = 0
+
+	def fit(self,x,y):
+		n = len(x)
+		m_x,m_y = np.mean(x),np.mean(y)
+		self.w_1 = np.sum(y * x - n * m_x * m_y) / np.sum(x * x - n * m_x * m_x)
+		self.w_0 = m_y - (self.w_1 * m_x)
+		return(self.w_0,self.w_1)
+
+	def predict(self,x_test):
+		y_pred = self.w_0 + self.w_1 * x_test
+		return y_pred
+
+	def plot(self,x,y):
+		plt.scatter(x,y,color="r",marker="o")
+		line=self.w_0 + self.w_1 * x
+		plt.plot(x,line,color="g")
+		plt.xlabel("X")
+		plt.ylabel("Y")
+		plt.show()
+
+if __name__ == "__main__":
+	x=np.array([1,2,3,4,5,6,7,8,9,10])
+	y=np.array([300,400,500,600,700,800,900,1000,1200,1400])
+	model=SimpleLinearRegression()
+	model.fit(x,y)
+	print(model.predict(11))
+	model.plot(x,y)