Contribution - Logistic Regression

Algorithm/LogisticRegression.py

@@ -0,0 +1,70 @@
+import numpy as np
+
+"""  
+Losgistic Regression is Nothing but
+creating probability distribution insted of specific values
+So that we can clearly classify between two classes
+
+
+Initailizing Weights and Bias as ZERO 
+and with the Iteration changing them
+"""
+
+
+class LogisticRegression:
+    def __init__(self, learning_rate=0.001, iter_count=1000):
+        """
+        Para
+        learning_rate = Learning Rate
+        iter_count = Number of Iterations (default value = 1000)
+        """
+        self.learning_rate = learning_rate
+        self.iter_count = iter_count
+        self.weights = None
+        self.bias = None
+
+    def my_sigmoid(z):
+        """
+        y^ = 1 / (1 + EXP(-z))
+        # sigmoid function for variable  (z)
+        """
+        result = 1 / (1 + np.exp(-z))
+        return result
+
+    def fit(self, X, y):
+        sample_count, feature_count = X.shape
+
+        self.weights = np.zeros(self.features)
+        self.bias = 0
+
+        # z = w.x+b
+        # y^ = 1 / (1 + EXP(-z))
+        # sigmoid function
+
+        for i in range(self.iter_count):
+            linear_prediction = np.dot(X, self.weights) + self.bias
+            prediction = LogisticRegression.my_sigmoid(linear_prediction)
+
+            # Calculating Error
+
+            # gradient for weights and bias
+            dw = (1 / sample_count) * np.dot(X.T, (prediction - y))
+            db = (1 / sample_count) * np.sum(prediction - y)
+
+            self.weights = self.weights - self.learning_rate * dw
+            self.bias = self.bias - self.learning_rate * db
+
+    def predict_proba(self, X_test):
+        linear_prediction = np.dot(X_test, self.weights) + self.bias
+        y_prediction = LogisticRegression.my_sigmoid(linear_prediction)
+
+        return y_prediction
+
+    def predict(self, X_test):
+        class_prediction = [0 if y <= 0.5 else 1 for y in self.predict_proba(X_test)]
+
+        return class_prediction
+
+    def accuracy(y_pred , y_test):
+
+        return np.sum(y_pred == y_test) / len(y_test)