Update app.py

app.py

@@ -0,0 +1,181 @@
+import streamlit as st
+import pandas as pd
+import numpy as np
+from sklearn.datasets import load_iris
+from sklearn.model_selection import train_test_split
+from sklearn.svm import SVC
+from sklearn.linear_model import LogisticRegression
+from sklearn.preprocessing import StandardScaler
+from sklearn.metrics import classification_report, accuracy_score
+import matplotlib.pyplot as plt
+import seaborn as sns
+from io import StringIO
+
+# Page config
+st.set_page_config(page_title="Explore SVM Algorithm", layout="wide")
+st.title("🔍 Support Vector Machine (SVM) Classifier Explained")
+
+# -----------------------------------
+# Theory Section
+# -----------------------------------
+st.markdown("""
+## 🤖 What is a Support Vector Machine (SVM)?
+
+SVM is a powerful supervised learning algorithm used for classification and regression.
+It works by finding a hyperplane that best separates the classes in the feature space.
+
+**Key Ideas:**
+- Maximizes the margin between different classes
+- Effective in high-dimensional spaces
+- Can use **kernel tricks** to handle non-linear classification
+
+---
+## ⚙️ How SVM Works
+
+1. Find the optimal hyperplane that separates classes.
+2. Use **support vectors** — data points closest to the hyperplane.
+3. Maximize the margin between these support vectors.
+4. Use **kernel functions** to map inputs to higher dimensions if data isn't linearly separable.
+
+**Kernel Types:**
+- *Linear*: Straight line separation  
+- *RBF (Gaussian)*: Circular, good for complex boundaries  
+- *Polynomial*: Curved boundaries  
+---
+""")
+
+# -----------------------------------
+# Load Dataset
+# -----------------------------------
+st.subheader("🌼 Try SVM on the Iris Dataset")
+iris = load_iris()
+df = pd.DataFrame(iris.data, columns=iris.feature_names)
+df['target'] = iris.target
+df['species'] = df['target'].apply(lambda x: iris.target_names[x])
+st.dataframe(df.head(), use_container_width=True)
+
+# -----------------------------------
+# Model Controls
+# -----------------------------------
+kernel = st.radio("Select SVM Kernel", ["linear", "rbf", "poly"])
+C = st.slider("Select Regularization Parameter (C)", 0.01, 10.0, value=1.0)
+
+# Prepare Data
+X = df.drop(columns=["target", "species"])
+y = df['target']
+
+scaler = StandardScaler()
+X_scaled = scaler.fit_transform(X)
+
+X_train, X_test, y_train, y_test = train_test_split(X_scaled, y, test_size=0.2, random_state=42)
+
+# Train SVM
+svm_model = SVC(kernel=kernel, C=C, probability=True, random_state=42)
+svm_model.fit(X_train, y_train)
+svm_pred = svm_model.predict(X_test)
+
+svm_acc = accuracy_score(y_test, svm_pred)
+st.success(f"✅ SVM Accuracy: {svm_acc*100:.2f}%")
+
+# -----------------------------------
+# Classification Report
+# -----------------------------------
+svm_report = classification_report(y_test, svm_pred, target_names=iris.target_names)
+st.markdown("### 📊 SVM Classification Report")
+st.text(svm_report)
+
+# -----------------------------------
+# Compare with Logistic Regression
+# -----------------------------------
+st.markdown("### 🔁 Compare with Logistic Regression")
+
+log_model = LogisticRegression(max_iter=200)
+log_model.fit(X_train, y_train)
+log_pred = log_model.predict(X_test)
+log_acc = accuracy_score(y_test, log_pred)
+
+st.info(f"📈 Logistic Regression Accuracy: {log_acc*100:.2f}%")
+
+if log_acc > svm_acc:
+    st.warning("🤔 Logistic Regression outperformed SVM! Try tuning SVM parameters or switching kernels.")
+else:
+    st.success("✅ SVM performed better than Logistic Regression on this dataset.")
+
+# -----------------------------------
+# Visualize Decision Boundaries
+# -----------------------------------
+st.markdown("### 🌌 Visualizing Decision Boundaries (2 Features)")
+
+feature_x = st.selectbox("Feature for X-axis", df.columns[:-2], index=0)
+feature_y = st.selectbox("Feature for Y-axis", df.columns[:-2], index=1)
+
+X_vis = df[[feature_x, feature_y]]
+X_vis_scaled = scaler.fit_transform(X_vis)
+X_train_v, X_test_v, y_train_v, y_test_v = train_test_split(X_vis_scaled, y, test_size=0.2, random_state=42)
+
+model_vis = SVC(kernel=kernel, C=C)
+model_vis.fit(X_train_v, y_train_v)
+
+h = .02
+x_min, x_max = X_vis_scaled[:, 0].min() - 1, X_vis_scaled[:, 0].max() + 1
+y_min, y_max = X_vis_scaled[:, 1].min() - 1, X_vis_scaled[:, 1].max() + 1
+xx, yy = np.meshgrid(np.arange(x_min, x_max, h), np.arange(y_min, y_max, h))
+Z = model_vis.predict(np.c_[xx.ravel(), yy.ravel()])
+Z = Z.reshape(xx.shape)
+
+fig, ax = plt.subplots(figsize=(8, 6))
+plt.contourf(xx, yy, Z, alpha=0.3)
+sns.scatterplot(x=X_vis_scaled[:, 0], y=X_vis_scaled[:, 1], hue=df['species'], palette='deep', ax=ax)
+plt.xlabel(feature_x)
+plt.ylabel(feature_y)
+plt.title("SVM Decision Boundaries")
+st.pyplot(fig)
+
+# -----------------------------------
+# Downloadable Report
+# -----------------------------------
+st.markdown("### 📥 Download SVM Report")
+st.download_button("📄 Download Classification Report", data=svm_report, file_name="svm_report.txt")
+
+# -----------------------------------
+# Summary
+# -----------------------------------
+st.markdown("""
+---
+## 💡 Highlights of SVM:
+- Works well for both linear and non-linear problems.
+- Excellent performance on small to medium-sized datasets.
+- Sensitive to outliers but tunable via regularization.
+
+## 🔧 When to Use SVM?
+Use them when:
+- You have a clear margin of separation between classes.
+- You're dealing with high-dimensional data.
+- You want flexibility via kernels.
+
+---
+### 🧠 Did You Know?
+
+- SVMs are **robust to overfitting**, especially in high-dimensional space.
+- The **'C' parameter** controls the trade-off between training error and margin size.
+- The **kernel trick** allows SVMs to operate in infinite-dimensional space.
+
+### 📌 Pros & Cons
+
+| Pros                            | Cons                                |
+|---------------------------------|-------------------------------------|
+| Works well on complex boundaries| Slower on large datasets            |
+| Effective in high-dimensional space | Needs careful parameter tuning |
+| Can handle non-linear data      | Less interpretable than simpler models |
+
+---
+### 🌀 Kernel Choice Summary
+
+| Kernel      | Use Case                        |
+|-------------|---------------------------------|
+| Linear      | Simple, linearly separable data |
+| RBF         | Most common, good for most cases|
+| Polynomial  | Use if you suspect curved boundaries|
+
+> 🎯 *Tip:* Start with linear, then try RBF if the data isn't linearly separable.
+""")