app.py

import streamlit as st
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.cluster import AgglomerativeClustering
from sklearn.metrics import confusion_matrix

# Load dataset
@st.cache_data
def load_data():
    file_path = "car_price_dataset.csv"  # Ensure this file is in the same directory
    return pd.read_csv(file_path)

df = load_data()

# Streamlit App Title
st.title("🚗 Car Price Clustering & Evaluation")

# Creating Tabs
tab1, tab2, tab3 = st.tabs(["📊 Dataset Overview", "📈 Visual Matrix", "⚙️ User Input for Clustering"])

# --- TAB 1: Dataset Overview ---
with tab1:
    st.write("## Dataset Overview")
    st.write(df.head())
    st.write(df.describe())

# --- TAB 2: Visualization Matrix ---
with tab2:
    st.write("## Data Visualization")

    # Select numerical features
    numerical_df = df.select_dtypes(include=[np.number])

    # Correlation Heatmap
    st.write("### Correlation Heatmap")
    fig, ax = plt.subplots(figsize=(8, 5))
    sns.heatmap(numerical_df.corr(), annot=True, cmap="coolwarm", fmt=".2f")
    st.pyplot(fig)

    # Confusion Matrix
    st.write("### Confusion Matrix")
    selected_features = ["Engine_Size", "Mileage", "Price"]
    if all(f in numerical_df.columns for f in selected_features):
        X = df[selected_features].dropna().values
        n_clusters = 3  # Default cluster count

        # Apply Hierarchical Clustering
        hc = AgglomerativeClustering(n_clusters=n_clusters, linkage='ward')
        labels = hc.fit_predict(X)

        # Generate dummy "true labels" (for demonstration)
        true_labels = np.random.randint(0, n_clusters, len(labels))

        cm = confusion_matrix(true_labels, labels)
        fig, ax = plt.subplots(figsize=(5, 4))
        sns.heatmap(cm, annot=True, cmap="Blues", fmt="d")
        plt.xlabel("Predicted")
        plt.ylabel("Actual")
        st.pyplot(fig)
    else:
        st.warning("Not enough numerical data for clustering.")

    # Scatter Plot
    st.write("### Scatter Plot")
    scatter_x = st.selectbox("Select X-axis", numerical_df.columns, index=0)
    scatter_y = st.selectbox("Select Y-axis", numerical_df.columns, index=1)

    fig, ax = plt.subplots(figsize=(6, 4))
    sns.scatterplot(x=df[scatter_x], y=df[scatter_y], alpha=0.7)
    plt.xlabel(scatter_x)
    plt.ylabel(scatter_y)
    st.pyplot(fig)

# --- TAB 3: User Input & Clustering ---
with tab3:
    st.write("## Perform Clustering")

    numerical_features = numerical_df.columns.tolist()
    selected_features = st.multiselect("Select features for clustering", numerical_features, default=["Engine_Size", "Mileage", "Price"])

    if len(selected_features) < 2:
        st.warning("Please select at least two numerical features.")
    else:
        X = df[selected_features].dropna().values  # Prepare data

        # Choose Number of Clusters (With + / - Buttons)
        n_clusters = st.number_input("Select Number of Clusters", min_value=2, max_value=10, value=3, step=1)

        # Predict Button
        if st.button("Predict Clusters"):
            # Apply Hierarchical Clustering
            hc = AgglomerativeClustering(n_clusters=n_clusters, linkage='ward')
            labels = hc.fit_predict(X)

            # Display results
            df["Cluster"] = labels
            st.write("### Clustered Data")
            st.write(df[selected_features + ["Cluster"]].head(10))

            st.success("Clustering Complete! 🎉")