venkyo8/visuals_With_Different_Algorithms

1import streamlit as st2import numpy as np3import matplotlib.pyplot as plt4from sklearn.datasets import make_classification, make_circles, make_blobs, make_moons5from sklearn.model_selection import train_test_split, learning_curve6from sklearn.neighbors import KNeighborsClassifier7from sklearn.naive_bayes import GaussianNB8from sklearn.linear_model import LogisticRegression9from sklearn.tree import DecisionTreeClassifier10from sklearn.ensemble import RandomForestClassifier11from mlxtend.plotting import plot_decision_regions12 13st.image(r"Innomatics-Logo.png", width=700)14st.title('visuals With Different Algorithms')15 16data = st.sidebar.selectbox('Type of Data', ('classification', 'circles', 'blobs', 'moons'))17 18if data == 'classification':19    X, y = make_classification(n_samples=250, n_features=2, n_informative=2, n_redundant=0, random_state=42)20elif data == 'circles':21    X, y = make_circles(n_samples=250, factor=0.5, noise=0.05)22elif data == 'blobs':23    X, y = make_blobs(n_samples=250, centers=2, n_features=2, cluster_std=1.0, random_state=42)24elif data == 'moons':25    X, y = make_moons(n_samples=250, noise=0.1, random_state=42)26 27X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.3, random_state=42)28 29def plot_decision_surface(X, y, model, title):30    plt.figure(figsize=(6, 4))31    plot_decision_regions(X, y, clf=model, colors="#7f7f7f,#bcbd22,#17becf")32    plt.title(title)33    st.pyplot(plt, clear_figure=True)34 35def plot_learning_curve(model, title):36    train_sizes, train_scores, test_scores = learning_curve(model, X_train, y_train, cv=5, scoring='accuracy')37    train_mean = np.mean(train_scores, axis=1)38    test_mean = np.mean(test_scores, axis=1)39    40    plt.figure(figsize=(6, 4))41    plt.plot(train_sizes, train_mean, label='Training Accuracy', marker='o')42    plt.plot(train_sizes, test_mean, label='Testing Accuracy', marker='s')43    plt.xlabel('Training Examples')44    plt.ylabel('Accuracy')45    plt.title(f'Learning Curve: {title}')46    plt.legend()47    st.pyplot(plt, clear_figure=True)48 49classifier_name = st.sidebar.selectbox('Select Classifier', ('KNN', 'Naive Bayes', 'Logistic Regression', 'Decision Tree', 'Random Forest'))50 51if classifier_name == 'KNN':52    n_neighbors = st.sidebar.slider('Number of Neighbors (k)', 1, 15, 3)53    knn = KNeighborsClassifier(n_neighbors=n_neighbors)54    knn.fit(X_train, y_train)55    plot_decision_surface(X, y, knn, 'KNeighbors Classifier')56    plot_learning_curve(knn, 'KNeighbors Classifier')57 58elif classifier_name == 'Naive Bayes':59    nb = GaussianNB()60    nb.fit(X_train, y_train)61    plot_decision_surface(X, y, nb, 'Naive Bayes')62    plot_learning_curve(nb, 'Naive Bayes')63 64elif classifier_name == 'Decision Tree':65    dt = DecisionTreeClassifier()66    dt.fit(X_train, y_train)67    plot_decision_surface(X, y, dt, 'Decision Tree')68    plot_learning_curve(dt, 'Decision Tree')69 70elif classifier_name == 'Random Forest':71    n_estimators = st.sidebar.slider('Number of Trees', 10, 200, 100)72    rf = RandomForestClassifier(n_estimators=n_estimators, random_state=42)73    rf.fit(X_train, y_train)74    plot_decision_surface(X, y, rf, 'Random Forest')75    plot_learning_curve(rf, 'Random Forest')76 77else:78    lr = LogisticRegression()79    lr.fit(X_train, y_train)80    plot_decision_surface(X, y, lr, 'Logistic Regression')81    plot_learning_curve(lr, 'Logistic Regression')82 83