#9 🧠 Machine Learning Dashboards in Streamlit (Regression + Classification + Charts)

Want to see your ML models in action with sliders, dropdowns, and live predictions? Streamlit makes that super easy — no frontend code, just Python! Let’s build interactive dashboards for both regression and classification tasks. 🚀

---

1️⃣ Install Libraries

!pip install streamlit scikit-learn plotly pandas numpy

---

2️⃣ Start Your Streamlit App

import streamlit as st
import pandas as pd
import numpy as np
from sklearn.model_selection import train_test_split
from sklearn.linear_model import LinearRegression, LogisticRegression
from sklearn.datasets import load_iris, fetch_california_housing
from sklearn.metrics import mean_squared_error, accuracy_score
import plotly.express as px

st.set_page_config(page_title="ML Dashboards", layout="wide")
st.title("🧠 Machine Learning Dashboard")
st.sidebar.header("⚙️ Choose Task")

task = st.sidebar.radio("Select Type:", ["Regression", "Classification"])

---

3️⃣ Regression Dashboard — California Housing

if task == "Regression":
    st.header("🏡 California Housing — Linear Regression")
    data = fetch_california_housing(as_frame=True)
    df = data.frame

    st.write("### Dataset Preview")
    st.dataframe(df.head())

    X = df.drop("MedHouseVal", axis=1)
    y = df["MedHouseVal"]
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

    model = LinearRegression()
    model.fit(X_train, y_train)
    preds = model.predict(X_test)
    mse = mean_squared_error(y_test, preds)

    st.metric("Mean Squared Error", round(mse, 3))

    fig = px.scatter(x=y_test, y=preds, labels={'x':'Actual', 'y':'Predicted'}, title="Actual vs Predicted Values")
    st.plotly_chart(fig, use_container_width=True)

---

4️⃣ Classification Dashboard — Iris Dataset

if task == "Classification":
    st.header("🌸 Iris Flower Classification — Logistic Regression")
    iris = load_iris(as_frame=True)
    df = iris.frame

    st.write("### Dataset Preview")
    st.dataframe(df.head())

    X = df.drop("target", axis=1)
    y = df["target"]
    X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)

    model = LogisticRegression(max_iter=500)
    model.fit(X_train, y_train)
    preds = model.predict(X_test)
    acc = accuracy_score(y_test, preds)

    st.metric("Accuracy", f"{acc*100:.2f}%")

    fig = px.scatter_3d(df, x="sepal length (cm)", y="sepal width (cm)", z="petal length (cm)",
                        color=df["target"].map({0:"Setosa",1:"Versicolor",2:"Virginica"}),
                        title="3D Flower Visualization")
    st.plotly_chart(fig, use_container_width=True)

---

5️⃣ Add Real-Time User Prediction

if task == "Classification":
    st.subheader("🔮 Predict Flower Type")

    sl = st.slider("Sepal Length (cm)", 4.0, 8.0, 5.8)
    sw = st.slider("Sepal Width (cm)", 2.0, 4.5, 3.0)
    pl = st.slider("Petal Length (cm)", 1.0, 7.0, 4.3)
    pw = st.slider("Petal Width (cm)", 0.1, 2.5, 1.3)

    sample = np.array([[sl, sw, pl, pw]])
    pred = model.predict(sample)[0]
    species = iris.target_names[pred]

    st.success(f"🌼 Predicted Flower: **{species}**")

---

6️⃣ Run the App

# Save the file as app.py
# Run this in terminal
streamlit run app.py  (or) python -m streamlit run app.py

Open the link Streamlit gives (usually http://localhost:8501) — and your ML dashboard is live 🎯

---

Part of Response(s) should be like 👇

Regression

Classification

✅ What You Learned

• ⚙️ Use Streamlit for real-time model dashboards
• 🏡 Build a regression dashboard with metrics & charts
• 🌸 Add interactive classification UI with sliders
• 📊 Visualize results using Plotly

Now bro, you can turn any ML model into a live web app in under 50 lines of Python 🔥 Next up? “Streamlit + LangChain for Agentic AI Apps” — want me to make that next?