Code Editor for app.py

import streamlit as st
import pandas as pd
import plotly.express as px
import seaborn as sns
from typing import Optional, Tuple, List

@st.cache_data
def load_data(url: str) -> Optional[pd.DataFrame]:
    """Load data from URL and cache it"""
    try:
        return pd.read_csv(url)
    except Exception as e:
        st.error(f"Error loading data: {str(e)}")
        return None

def plot_data(df: pd.DataFrame, variables: List[str]) -> None:
    """Plot the selected variables using the most appropriate visualization"""
    if len(variables) == 1:
        var = variables[0]
        if df[var].dtype in ['int64', 'float64']:
            # Numerical variable - show histogram
            fig = px.histogram(df, x=var, title=f"Distribution of {var}")
            st.plotly_chart(fig)
        else:
            # Categorical variable - show count plot
            fig = px.bar(df[var].value_counts().reset_index(), 
                        x='index', y=var,
                        title=f"Count plot of {var}")
            st.plotly_chart(fig)
    
    elif len(variables) == 2:
        var1, var2 = variables
        if df[var1].dtype in ['int64', 'float64'] and df[var2].dtype in ['int64', 'float64']:
            # Both numerical - scatter plot
            fig = px.scatter(df, x=var1, y=var2, title=f"{var1} vs {var2}")
            st.plotly_chart(fig)
        elif df[var1].dtype in ['int64', 'float64'] or df[var2].dtype in ['int64', 'float64']:
            # One numerical, one categorical - box plot
            num_var = var1 if df[var1].dtype in ['int64', 'float64'] else var2
            cat_var = var2 if df[var1].dtype in ['int64', 'float64'] else var1
            fig = px.box(df, x=cat_var, y=num_var, title=f"{num_var} by {cat_var}")
            st.plotly_chart(fig)
        else:
            # Both categorical - heatmap
            contingency = pd.crosstab(df[var1], df[var2])
            fig = px.imshow(contingency, 
                          title=f"Relationship between {var1} and {var2}",
                          labels=dict(x=var2, y=var1, color="Count"))
            st.plotly_chart(fig)

# Main app
st.title("Data Explorer")

# URL input
url = st.text_input(
    "Enter CSV URL",
    value="https://raw.githubusercontent.com/mwaskom/seaborn-data/master/penguins.csv"
)

# Load data
if url:
    df = load_data(url)
    
    if df is not None:
        # Controls
        col1, col2 = st.columns(2)
        
        with col1:
            n_rows = st.number_input("Number of rows to display", 
                                   min_value=1, 
                                   max_value=len(df), 
                                   value=min(5, len(df)))
        
        with col2:
            shuffle = st.checkbox("Randomly shuffle data")
        
        # Display data
        if shuffle:
            display_df = df.sample(frac=1).reset_index(drop=True)
        else:
            display_df = df
            
        st.dataframe(display_df.head(n_rows))
        
        # Variable selection
        st.subheader("Visualization")
        cols = df.columns.tolist()
        vars_to_plot = st.multiselect(
            "Select 1 or 2 variables to visualize",
            options=cols,
            max_selections=2
        )
        
        if vars_to_plot:
            plot_data(df, vars_to_plot)