dthota10/Different_Machine_Learning_Algorithms

1import streamlit as st
2import numpy as np
3import matplotlib.pyplot as plt
4from sklearn.datasets import make_classification, make_circles, make_blobs, make_moons
5from sklearn.model_selection import train_test_split, learning_curve
6from sklearn.neighbors import KNeighborsClassifier
7from sklearn.naive_bayes import GaussianNB
8from sklearn.linear_model import LogisticRegression
9from sklearn.tree import DecisionTreeClassifier
10from sklearn.metrics import accuracy_score, f1_score
11from mlxtend.plotting import plot_decision_regions
12
13# Display image
14st.image("inno.jpg", width=600)
15
16# Streamlit app title
17st.markdown("<h1 style='color:#2ca02c;'>Visualizing Classification Boundaries</h1>", unsafe_allow_html=True)
18
19# Select dataset
20data = st.sidebar.selectbox('Type of data ', ('Classification', 'Circles', 'Blobs', 'Moons'))
21
22if data == 'Classification':
23    X, y = make_classification(n_samples=300, n_features=2, n_redundant=0, random_state=42)
24elif data == 'Circles':
25    X, y = make_circles(n_samples=300, factor=0.5, noise=0.05)
26elif data == 'Blobs':
27    X, y = make_blobs(n_samples=300, centers=2, n_features=2, cluster_std=1.0, random_state=42)
28elif data == 'Moons':
29    X, y = make_moons(n_samples=300, noise=0.1, random_state=42)
30
31# Split dataset
32X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)
33
34def plot_decision_surface(X, y, model, title):
35    plt.figure(figsize=(6,4))
36    plot_decision_regions(X, y, clf=model, colors="#98df8a")
37    plt.title(title)
38    st.pyplot(plt.gcf(), clear_figure=True)
39
40# Select classifier
41classifier_name = st.sidebar.selectbox('Select Classifier', ('KNN', 'Naive Bayes', 'Logistic Regression', 'DecisionTreeClassifier'))
42
43if classifier_name == 'KNN':
44    n_neighbors = st.sidebar.slider('Number of Neighbors (k)', 1, 15, 3)
45    weights = st.sidebar.radio('Weight Function', ('uniform', 'distance'))
46    algorithm = st.sidebar.selectbox('Algorithm', ('auto', 'ball_tree', 'kd_tree', 'brute'))
47    
48    model = KNeighborsClassifier(n_neighbors=n_neighbors, weights=weights, algorithm=algorithm)
49    
50elif classifier_name == 'Naive Bayes':
51    model = GaussianNB()
52    
53elif classifier_name == 'DecisionTreeClassifier':
54    model = DecisionTreeClassifier()
55    
56else:
57    model = LogisticRegression()
58
59# Train model
60model.fit(X_train, y_train)
61
62# Make predictions
63y_pred = model.predict(X_test)
64
65# Compute accuracy & F1-score
66accuracy = accuracy_score(y_test, y_pred)
67f1 = f1_score(y_test, y_pred)
68
69# Display metrics in Streamlit
70st.subheader("Model Performance")
71st.write(f"Accuracy: {accuracy:.2f}")
72st.write(f"F1-score: {f1:.2f}")
73
74# Plot decision boundary
75plot_decision_surface(X, y, model, f'{classifier_name} Decision Surface')
76
77# Plot Learning Curve
78def plot_learning_curve(model, X, y):
79    train_sizes, train_scores, test_scores = learning_curve(model, X, y, cv=5, scoring='accuracy', train_sizes=np.linspace(0.1, 1.0, 10))
80    
81    train_mean = np.mean(train_scores, axis=1)
82    test_mean = np.mean(test_scores, axis=1)
83    
84    plt.figure(figsize=(6,4))
85    plt.plot(train_sizes, train_mean, 'o-', label="Training Accuracy")
86    plt.plot(train_sizes, test_mean, 'o-', label="Validation Accuracy")
87    
88    plt.xlabel("Training Samples")
89    plt.ylabel("Accuracy")
90    plt.title(f"Learning Curve: {classifier_name}")
91    plt.legend()
92    st.pyplot(plt.gcf(), clear_figure=True)
93
94# Display Learning Curve
95st.subheader("Learning Curve")
96plot_learning_curve(model, X, y)