From ade3ed3784fe8e34978563df76a7b4f7d5fe3665 Mon Sep 17 00:00:00 2001
From: Taru <32308101+epicalyx@users.noreply.github.com>
Date: Wed, 17 Oct 2018 00:22:44 +0530
Subject: [PATCH 1/3] Logistic regression implementation

implementation of logistic regression for binary classification
---
 machine_learning/logistic_regression.py | 97 +++++++++++++++++++++++++
 1 file changed, 97 insertions(+)
 create mode 100644 machine_learning/logistic_regression.py

diff --git a/machine_learning/logistic_regression.py b/machine_learning/logistic_regression.py
new file mode 100644
index 000000000..70c0b2807
--- /dev/null
+++ b/machine_learning/logistic_regression.py
@@ -0,0 +1,97 @@
+#!/usr/bin/env python
+# coding: utf-8
+
+# # Logistic Regression from scratch
+
+# In[62]:
+
+
+''' Implementing logistic regression for classification problem 
+     Helpful resources : 1.Coursera ML course    2.https://medium.com/@martinpella/logistic-regression-from-scratch-in-python-124c5636b8ac'''
+
+
+# In[63]:
+
+
+#importing all the required libraries
+import numpy as np
+import matplotlib.pyplot as plt
+get_ipython().run_line_magic('matplotlib', 'inline')
+from sklearn import datasets
+
+
+# In[67]:
+
+
+#sigmoid function or logistic function is used as a hypothesis function in classification problems
+def sigmoid_function(z):
+    return 1/(1+np.exp(-z))
+
+
+def cost_function(h,y):
+    return (-y*np.log(h)-(1-y)*np.log(1-h)).mean()
+
+# here alpha is the learning rate, X is the featue matrix,y is the target matrix
+def logistic_reg(alpha,X,y,max_iterations=70000):
+    converged=False
+    iterations=0
+    theta=np.zeros(X.shape[1])
+    
+    num_iterations=0
+    while not converged:
+        z=np.dot(X,theta)
+        h=sigmoid_function(z)
+        gradient = np.dot(X.T,(h-y))/y.size
+        theta=theta-(alpha)*gradient
+        
+        z=np.dot(X,theta)
+        h=sigmoid_function(z)
+        e=cost_function(h,y)
+        print('J=',e)
+        J=e
+        
+        iterations+=1   #update iterations
+        
+        
+        if iterations== max_iterations:
+            print("Maximum iterations exceeded!")
+            converged=True
+            
+    return theta
+
+
+
+        
+    
+    
+
+
+# In[68]:
+
+
+if __name__=='__main__':
+    iris=datasets.load_iris()
+    X = iris.data[:, :2]
+    y = (iris.target != 0) * 1
+    
+    alpha=0.1
+    theta=logistic_reg(alpha,X,y,max_iterations=70000)
+    print(theta)
+    def predict_prob(X):
+        return sigmoid_function(np.dot(X,theta))              # predicting the value of probability from the logistic regression algorithm
+        
+    
+    plt.figure(figsize=(10, 6))
+    plt.scatter(X[y == 0][:, 0], X[y == 0][:, 1], color='b', label='0')
+    plt.scatter(X[y == 1][:, 0], X[y == 1][:, 1], color='r', label='1')
+    x1_min, x1_max = X[:,0].min(), X[:,0].max(),
+    x2_min, x2_max = X[:,1].min(), X[:,1].max(),
+    xx1, xx2 = np.meshgrid(np.linspace(x1_min, x1_max), np.linspace(x2_min, x2_max))
+    grid = np.c_[xx1.ravel(), xx2.ravel()]
+    probs = predict_prob(grid).reshape(xx1.shape)
+    plt.contour(xx1, xx2, probs, [0.5], linewidths=1, colors='black');
+
+    plt.legend();
+    
+    
+

From f018ddc4c00f71e7123b780d409550e101e3c673 Mon Sep 17 00:00:00 2001
From: Taru <32308101+epicalyx@users.noreply.github.com>
Date: Wed, 17 Oct 2018 00:52:32 +0530
Subject: [PATCH 2/3] requested changes addressed

---
 machine_learning/logistic_regression.py | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)

diff --git a/machine_learning/logistic_regression.py b/machine_learning/logistic_regression.py
index 70c0b2807..2e28fd96b 100644
--- a/machine_learning/logistic_regression.py
+++ b/machine_learning/logistic_regression.py
@@ -31,13 +31,13 @@ def sigmoid_function(z):
 def cost_function(h,y):
     return (-y*np.log(h)-(1-y)*np.log(1-h)).mean()
 
-# here alpha is the learning rate, X is the featue matrix,y is the target matrix
+# here alpha is the learning rate, X is the feature matrix,y is the target matrix
 def logistic_reg(alpha,X,y,max_iterations=70000):
     converged=False
     iterations=0
     theta=np.zeros(X.shape[1])
     
-    num_iterations=0
+    
     while not converged:
         z=np.dot(X,theta)
         h=sigmoid_function(z)
@@ -46,9 +46,9 @@ def logistic_reg(alpha,X,y,max_iterations=70000):
         
         z=np.dot(X,theta)
         h=sigmoid_function(z)
-        e=cost_function(h,y)
-        print('J=',e)
-        J=e
+        J=cost_function(h,y)
+       
+     
         
         iterations+=1   #update iterations
         

From 7105f6f648aa728c44e04e2389ad141314e2818f Mon Sep 17 00:00:00 2001
From: Taru <32308101+epicalyx@users.noreply.github.com>
Date: Wed, 17 Oct 2018 01:07:29 +0530
Subject: [PATCH 3/3] minor changes

requested changes are addressed
---
 machine_learning/logistic_regression.py | 1 +
 1 file changed, 1 insertion(+)

diff --git a/machine_learning/logistic_regression.py b/machine_learning/logistic_regression.py
index 2e28fd96b..de0cfd54d 100644
--- a/machine_learning/logistic_regression.py
+++ b/machine_learning/logistic_regression.py
@@ -55,6 +55,7 @@ def logistic_reg(alpha,X,y,max_iterations=70000):
         
         if iterations== max_iterations:
             print("Maximum iterations exceeded!")
+            print("Minimal cost function J=",J)
             converged=True
             
     return theta