From 657eedac922ec3446dfa5a9f70542ebf611982f8 Mon Sep 17 00:00:00 2001
From: Ritik Tiwari <111584498+testgithubtiwari@users.noreply.github.com>
Date: Sat, 24 Dec 2022 09:15:05 +0530
Subject: [PATCH] LDA code has been uploaded.

---
 untitled14.py | 82 +++++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 82 insertions(+)
 create mode 100644 untitled14.py

diff --git a/untitled14.py b/untitled14.py
new file mode 100644
index 0000000..55bf455
--- /dev/null
+++ b/untitled14.py
@@ -0,0 +1,82 @@
+# -*- coding: utf-8 -*-
+"""Untitled14.ipynb
+
+Automatically generated by Colaboratory.
+
+Original file is located at
+    https://colab.research.google.com/drive/1tCT28q7D3YlMnGmVLEiFWsAdT2hlW-ZY
+
+This is from library of LDA
+"""
+
+import matplotlib.pyplot as plt  #importing pyplot from  matplotlib 
+from sklearn import datasets  
+from sklearn.discriminant_analysis import LinearDiscriminantAnalysis  #applying LDA
+
+iris = datasets.load_iris()  
+colors = ["navy", "turquoise", "darkorange"]
+X = iris.data
+y = iris.target
+target_names = iris.target_names
+lda = LinearDiscriminantAnalysis(n_components=2)
+X_r2 = lda.fit(X, y).transform(X)
+plt.figure()
+for color, i, target_name in zip(colors, [0, 1, 2], target_names):
+    plt.scatter(
+        X_r2[y == i, 0], X_r2[y == i, 1], alpha=0.8, color=color, label=target_name
+    )
+plt.legend(loc="best", shadow=False, scatterpoints=1)
+plt.title("LDA of IRIS dataset")
+
+plt.show()
+
+"""This is from scratch"""
+
+import numpy as np
+import matplotlib.pyplot as plt
+from sklearn.model_selection import train_test_split
+from sklearn.datasets import load_iris
+
+class LDA:
+    def __init__(self, n_components=None):
+        self.n_components = n_components
+        self.eig_vectors = None
+    
+    def transform(self,X,y):
+        height, width = X.shape
+        unique_classes = np.unique(y)
+        num_classes = len(unique_classes)
+
+        scatter_t = np.cov(X.T)*(height - 1)
+        scatter_w = 0
+        for i in range(num_classes):
+            class_items = np.flatnonzero(y == unique_classes[i])
+            scatter_w = scatter_w + np.cov(X[class_items].T) * (len(class_items)-1)
+        
+        scatter_b = scatter_t - scatter_w
+        _, eig_vectors = np.linalg.eigh(np.linalg.pinv(scatter_w).dot(scatter_b))
+        print(eig_vectors.shape)
+        pc = X.dot(eig_vectors[:,::-1][:,:self.n_components])
+        print(pc.shape)
+
+        if self.n_components == 2:
+            if y is None:
+                plt.scatter(pc[:,0],pc[:,1])
+            else:
+                colors = ["navy", "turquoise", "darkorange"]
+                labels = np.unique(y)
+                for color, label in zip(colors, labels):
+                    class_data = pc[np.flatnonzero(y==label)]
+                    plt.scatter(class_data[:,0],class_data[:,1],c=color)
+            plt.show()
+        return pc
+
+LDA_obj = LDA(n_components=5)
+data = load_iris()
+X, y = data.data, data.target
+X_train, X_test, Y_train, Y_test = train_test_split(X, y, test_size=0.2)
+
+LDA_object = LDA(n_components=2)
+X_train_modified = LDA_object.transform(X_train, Y_train)
+
+print("Original Data Size:",X_train.shape, "\nModified Data Size:", X_train_modified.shape)
\ No newline at end of file