Py.Cafe

# check out https://dash.plotly.com/ for documentation
# And check out https://py.cafe/maartenbreddels for more examples
from dash import Dash, Input, Output, callback, dcc, html
import pandas as pd
import plotly.express as px
import plotly.graph_objects as go
import dash_bootstrap_components as dbc

# Load and prepare data
df = pd.read_csv('risky_habits.csv')
df.columns = df.columns.str.strip()

# Define custom ordering for categorical variables
income_order = [
    '$0 - $24,999',
    '$25,000 - $49,999', 
    '$50,000 - $99,999',
    '$100,000 - $149,999',
    '$150,000+'
]

education_order = [
    'Less than high school degree',
    'High school degree',
    'Some college or Associate degree',
    'Bachelor degree',
    'Graduate degree'
]

# Apply custom ordering
df['Household_Income'] = pd.Categorical(df['Household_Income'], categories=income_order, ordered=True)
df['Education'] = pd.Categorical(df['Education'], categories=education_order, ordered=True)

# Define all risk habits and characteristics
all_risk_habits = [
    'Do_you_ever_smoke_cigarettes',
    'Do_you_ever_drink_alcohol', 
    'Do_you_ever_gamble',
    'Have_you_ever_been_skydiving',
    'Do_you_ever_drive_above_the_speed_limit',
    'Have_you_ever_cheated_on_your_significant_other',
    'Do_you_eat_steak'
]

all_characteristics = [
    'Gender',
    'Age', 
    'Household_Income',
    'Education',
    'Location'
]

# Create nice display names for dropdowns
habit_options = []
for habit in all_risk_habits:
    # Clean up names for display
    display_name = habit.replace('Do_you_ever_', '').replace('Have_you_ever_', '')
    display_name = display_name.replace('_', ' ').title()
    habit_options.append({'label': display_name, 'value': habit})

char_options = []
for char in all_characteristics:
    display_name = char.replace('_', ' ').title()
    char_options.append({'label': display_name, 'value': char})

# Initialize Dash app
app = Dash(__name__, external_stylesheets=[dbc.themes.BOOTSTRAP])

# Define app layout
app.layout = dbc.Container([
    dbc.Row([
        dbc.Col([
            html.H1("Risky Habits Analysis Dashboard", className="text-center mb-4"),
            html.Hr()
        ])
    ]),
    
    # Risk Habits dropdown - full width, centered
    dbc.Row([
        dbc.Col([
            html.Label("Select Risk Habits:", className="fw-bold text-center d-block mb-2"),
            dcc.Dropdown(
                id='habit-dropdown',
                options=habit_options,
                value=all_risk_habits[:3],  # Default to first 3
                multi=True,
                placeholder="Select one or more risk habits...",
                style={'width': '100%'}
            )
        ], width="auto", className="mx-auto")
    ], className="mb-3"),
    
    # Characteristics dropdown - centered
    dbc.Row([
        dbc.Col([
            html.Label("Select Characteristic:", className="fw-bold text-center d-block mb-2"),
            dcc.Dropdown(
                id='char-dropdown',
                options=char_options,
                value='Gender',  # Default to Gender
                multi=False,
                placeholder="Select a characteristic...",
                style={'width': '100%'} 
            )
        ], width="auto", className="mx-auto")
    ], className="mb-4"),
    
    # Chart section with padding
    dbc.Row([
        dbc.Col([
            dcc.Graph(id='risk-chart')
        ], width=10, className="mx-auto")
    ], className="mb-3"),
    
    # Summary section - centered
    dbc.Row([
        dbc.Col([
            html.Div(id='data-summary', className="text-center")
        ], width=8, className="mx-auto")
    ])
    
], fluid=True, style={'backgroundColor': '#9cabb8', 'minHeight': '100vh', 'padding': '20px'})

# Callback for updating the chart
@app.callback(
    [Output('risk-chart', 'figure'),
     Output('data-summary', 'children')],
    [Input('habit-dropdown', 'value'),
     Input('char-dropdown', 'value')]
)
def update_chart(selected_habits, selected_char):
    # Error handling
    if not selected_habits or not selected_char:
        empty_fig = go.Figure()
        empty_fig.add_annotation(
            text="Please select at least one risk habit and one characteristic",
            xref="paper", yref="paper", x=0.5, y=0.5,
            showarrow=False, font=dict(size=16)
        )
        return empty_fig, "No data to display"
    
    try:
        # Check if columns exist
        missing_cols = []
        for habit in selected_habits:
            if habit not in df.columns:
                missing_cols.append(habit)
        if selected_char not in df.columns:
            missing_cols.append(selected_char)
            
        if missing_cols:
            error_fig = go.Figure()
            error_fig.add_annotation(
                text=f"Missing columns: {missing_cols}",
                xref="paper", yref="paper", x=0.5, y=0.5,
                showarrow=False, font=dict(size=16, color="red")
            )
            return error_fig, f"Error: Missing columns {missing_cols}"
        
        # Filter data for 'Yes' responses
        mask = df[selected_habits].eq('Yes').any(axis=1)
        filtered_df = df[mask]
        
        if len(filtered_df) == 0:
            empty_fig = go.Figure()
            empty_fig.add_annotation(
                text="No 'Yes' responses found for selected habits",
                xref="paper", yref="paper", x=0.5, y=0.5,
                showarrow=False, font=dict(size=16)
            )
            return empty_fig, "No 'Yes' responses found"
        
        # Count 'Yes' responses by characteristic
        risk_counts = filtered_df.groupby(selected_char)[selected_habits].apply(
            lambda x: (x == 'Yes').sum()
        ).reset_index()
        
        # Melt for plotting
        risk_melted = risk_counts.melt(
            id_vars=selected_char,
            var_name='Risk_Habit',
            value_name='Count'
        )
        
        # Clean up habit names for display
        risk_melted['Risk_Habit_Display'] = risk_melted['Risk_Habit'].str.replace('Do_you_ever_', '', regex=False)
        risk_melted['Risk_Habit_Display'] = risk_melted['Risk_Habit_Display'].str.replace('Have_you_ever_', '', regex=False)
        risk_melted['Risk_Habit_Display'] = risk_melted['Risk_Habit_Display'].str.replace('_', ' ', regex=False)
        risk_melted['Risk_Habit_Display'] = risk_melted['Risk_Habit_Display'].str.title()
        
        # Create grouped bar chart
        fig = px.bar(
            risk_melted,
            x=selected_char,
            y='Count',
            color='Risk_Habit_Display',
            barmode='group',
            title=f"Risk Habits by {selected_char.replace('_', ' ').title()}",
            labels={
                'Count': 'Number of "Yes" Responses',
                selected_char: selected_char.replace('_', ' ').title(),
                'Risk_Habit_Display': 'Risk Habit'
            },
            color_discrete_sequence=px.colors.qualitative.Dark2
        )
        
        fig.update_layout(
            height=500,
            xaxis_title=selected_char.replace('_', ' ').title(),
            yaxis_title='Number of "Yes" Responses',
            legend_title="Risk Habits",
            font=dict(size=12),
            title_font_size=16
        )
        
        # Create summary text
        total_responses = len(filtered_df)
        unique_chars = filtered_df[selected_char].nunique()
        summary = html.Div([
            html.P(f"Total respondents with at least one 'Yes': {total_responses}"),
            html.P(f"Number of {selected_char.replace('_', ' ').lower()} categories: {unique_chars}"),
            html.P(f"Analyzing {len(selected_habits)} risk habit(s)")
        ])
        
        return fig, summary
        
    except Exception as e:
        error_fig = go.Figure()
        error_fig.add_annotation(
            text=f"Error: {str(e)}",
            xref="paper", yref="paper", x=0.5, y=0.5,
            showarrow=False, font=dict(size=16, color="red")
        )
        return error_fig, f"Error occurred: {str(e)}"

# Run the app
if __name__ == '__main__':
    app.run(debug=True)