📊 Build an EDA Dashboard with Streamlit and Plotly

Streamlit turns Python scripts into web apps. The file_uploader widget lets users drop in any CSV.

# app.py
import streamlit as st
import pandas as pd
import plotly.express as px
import plotly.graph_objects as go

st.set_page_config(page_title='EDA Dashboard', layout='wide')

st.title('📊 EDA Dashboard')
st.caption('Upload any CSV file to explore your data with interactive charts.')

uploaded_file = st.file_uploader('Upload a CSV file', type=['csv'])

if uploaded_file:
    df = pd.read_csv(uploaded_file)
    st.success(f'Loaded {len(df):,} rows × {len(df.columns)} columns')
else:
    # Use a sample dataset if no file uploaded
    from sklearn.datasets import load_iris
    iris = load_iris(as_frame=True)
    df = iris.frame
    df['species'] = df['target'].map({0: 'setosa', 1: 'versicolor', 2: 'virginica'})
    st.info('No file uploaded — showing Iris dataset as demo. Upload your own CSV above.')

Display the first few rows, shape, data types, and descriptive statistics.

# Add this after loading df

# Dataset overview
col1, col2, col3 = st.columns(3)
col1.metric('Rows',    f'{len(df):,}')
col2.metric('Columns', len(df.columns))
col3.metric('Missing', f'{df.isnull().sum().sum():,}')

st.subheader('Preview')
st.dataframe(df.head(10), use_container_width=True)

st.subheader('Summary Statistics')
st.dataframe(df.describe(), use_container_width=True)

st.subheader('Data Types')
dtype_df = pd.DataFrame({
    'Column': df.dtypes.index,
    'Type':   df.dtypes.values.astype(str),
    'Missing': df.isnull().sum().values
})
st.dataframe(dtype_df, use_container_width=True)

st.subheader('Distribution — Histogram')

numeric_cols = df.select_dtypes(include='number').columns.tolist()
cat_cols     = df.select_dtypes(include='object').columns.tolist()

col_a, col_b = st.columns([2, 1])
with col_a:
    hist_col = st.selectbox('Select column', numeric_cols, key='hist_col')
with col_b:
    bins = st.slider('Number of bins', 10, 100, 30)

color_by = st.selectbox('Colour by (optional)', ['None'] + cat_cols, key='hist_color')
color_col = None if color_by == 'None' else color_by

fig_hist = px.histogram(
    df, x=hist_col, nbins=bins, color=color_col,
    title=f'Distribution of {hist_col}',
    template='plotly_white'
)
st.plotly_chart(fig_hist, use_container_width=True)

st.subheader('Correlation Heatmap')

if len(numeric_cols) >= 2:
    corr = df[numeric_cols].corr()

    fig_heat = go.Figure(data=go.Heatmap(
        z=corr.values,
        x=corr.columns.tolist(),
        y=corr.index.tolist(),
        colorscale='RdBu_r',
        zmid=0,
        text=corr.round(2).values,
        texttemplate='%{text}',
        textfont={'size': 11}
    ))
    fig_heat.update_layout(
        title='Pearson Correlation Matrix',
        template='plotly_white',
        height=500
    )
    st.plotly_chart(fig_heat, use_container_width=True)
else:
    st.info('Need at least 2 numeric columns for a correlation heatmap.')

st.subheader('Scatter Plot')

if len(numeric_cols) >= 2:
    col1, col2, col3 = st.columns(3)
    with col1:
        x_col = st.selectbox('X axis', numeric_cols, key='scatter_x')
    with col2:
        y_col = st.selectbox('Y axis', numeric_cols, index=1 if len(numeric_cols) > 1 else 0, key='scatter_y')
    with col3:
        color_col = st.selectbox('Colour by', ['None'] + cat_cols + numeric_cols, key='scatter_color')

    color_arg = None if color_col == 'None' else color_col

    fig_scatter = px.scatter(
        df, x=x_col, y=y_col, color=color_arg,
        title=f'{x_col} vs {y_col}',
        template='plotly_white',
        opacity=0.7,
        trendline='ols' if color_arg is None else None
    )
    st.plotly_chart(fig_scatter, use_container_width=True)

[Browser shows interactive Streamlit dashboard with:]
- 3 metric cards: Rows: 150 | Columns: 6 | Missing: 0
- Preview table with first 10 rows
- Summary statistics table
- Dropdown to pick histogram column + slider for bins
- Colour heatmap showing correlations between all numeric columns
- Scatter plot with X/Y/colour axis selectors