src/streamlit_app.py

import streamlit as st
import pandas as pd
import numpy as np
import joblib
import matplotlib.pyplot as plt
import seaborn as sns

from sklearn.model_selection import train_test_split
from sklearn.pipeline import Pipeline
from sklearn.compose import ColumnTransformer
from sklearn.preprocessing import OneHotEncoder, StandardScaler
from sklearn.ensemble import RandomForestClassifier
from sklearn.metrics import accuracy_score, confusion_matrix

st.set_page_config(page_title="Lead Conversion Predictor", layout="wide")

st.title("📊 ExtraaLearn Lead Conversion Predictor")

# Upload dataset
uploaded_file = st.file_uploader("Upload CSV Dataset", type=["csv"])

if uploaded_file:
    data = pd.read_csv(uploaded_file)
    st.subheader("Dataset Preview")
    st.dataframe(data.head())

    # Drop ID if exists
    if "ID" in data.columns:
        data = data.drop("ID", axis=1)

    # Target
    target = "status"

    # Split features
    X = data.drop(target, axis=1)
    y = data[target]

    # Identify columns
    cat_cols = X.select_dtypes(include="object").columns.tolist()
    num_cols = X.select_dtypes(exclude="object").columns.tolist()

    # Preprocessing
    preprocessor = ColumnTransformer([
        ("num", StandardScaler(), num_cols),
        ("cat", OneHotEncoder(handle_unknown="ignore"), cat_cols)
    ])

    # Model
    model = Pipeline([
        ("preprocessor", preprocessor),
        ("classifier", RandomForestClassifier(random_state=42))
    ])

    # Train-test split
    X_train, X_test, y_train, y_test = train_test_split(
        X, y, test_size=0.2, random_state=42
    )

    model.fit(X_train, y_train)

    preds = model.predict(X_test)
    acc = accuracy_score(y_test, preds)

    st.success(f"Model Accuracy: {acc:.2f}")

    # Confusion Matrix
    st.subheader("Confusion Matrix")
    cm = confusion_matrix(y_test, preds)

    fig, ax = plt.subplots()
    sns.heatmap(cm, annot=True, fmt="d", ax=ax)
    st.pyplot(fig)

    # ---------------------------
    # EDA SECTION
    # ---------------------------
    st.subheader("📈 Exploratory Data Analysis")

    col1, col2 = st.columns(2)

    with col1:
        st.write("### Age Distribution")
        fig, ax = plt.subplots()
        sns.histplot(data["age"], kde=True, ax=ax)
        st.pyplot(fig)

    with col2:
        st.write("### Website Visits")
        fig, ax = plt.subplots()
        sns.histplot(data["website_visits"], kde=True, ax=ax)
        st.pyplot(fig)

    st.write("### Conversion by Occupation")
    fig, ax = plt.subplots()
    sns.countplot(data=data, x="current_occupation", hue="status", ax=ax)
    st.pyplot(fig)

    # ---------------------------
    # PREDICTION UI
    # ---------------------------
    st.subheader("🎯 Predict New Lead")

    input_data = {}

    for col in X.columns:
        if col in cat_cols:
            input_data[col] = st.selectbox(col, data[col].unique())
        else:
            input_data[col] = st.number_input(col, float(data[col].min()), float(data[col].max()))

    if st.button("Predict"):
        input_df = pd.DataFrame([input_data])
        prediction = model.predict(input_df)[0]
        prob = model.predict_proba(input_df)[0][1]

        st.success(f"Prediction: {'Converted' if prediction == 1 else 'Not Converted'}")
        st.info(f"Conversion Probability: {prob:.2f}")