Create app.py

app.py

@@ -0,0 +1,295 @@
+import pandas as pd
+import plotly.graph_objects as go
+from plotly.subplots import make_subplots
+import gradio as gr
+import ast
+from functools import lru_cache
+from collections import Counter
+
+# --- Constants and Mappings (Unchanged) ---
+BODY_ORDER = ['Very light-bodied', 'Light-bodied', 'Medium-bodied', 'Full-bodied', 'Very full-bodied']
+ACIDITY_ORDER = ['Low', 'Medium', 'High']
+BODY_MAPPING = {'Very light-bodied': 1, 'Light-bodied': 2, 'Medium-bodied': 3, 'Full-bodied': 4, 'Very full-bodied': 5}
+WINE_TYPE_ORDER = {'Red': 2, 'Rosé': 1, 'White': 0}
+SAMPLE_THRESHOLDS = {'Common (100+)': 100, 'Uncommon (50+)': 50, 'Rare (20+)': 20}
+COUNTRY_FLAGS = {
+    'United States': '🇺🇸', 'France': '🇫🇷', 'Italy': '🇮🇹', 'Spain': '🇪🇸', 'Germany': '🇩🇪', 'Australia': '🇦🇺',
+    'Chile': '🇨🇱', 'Argentina': '🇦🇷', 'Portugal': '🇵🇹', 'South Africa': '🇿🇦', 'New Zealand': '🇳🇿',
+    'Austria': '🇦🇹', 'Greece': '🇬🇷', 'Hungary': '🇭🇺', 'Croatia': '🇭🇷', 'Slovenia': '🇸🇮', 'Canada': '🇨🇦',
+    'Brazil': '🇧🇷', 'Uruguay': '🇺🇾', 'Israel': '🇮🇱', 'Lebanon': '🇱🇧', 'Turkey': '🇹🇷', 'Bulgaria': '🇧🇬',
+    'Romania': '🇷🇴', 'Georgia': '🇬🇪', 'Moldova': '🇲🇩', 'Switzerland': '🇨🇭', 'England': '🏴'
+}
+FOOD_EMOJIS = {
+    'Beef': '🥩', 'Pork': '🍖', 'Lamb': '🍖', 'Poultry': '🍗', 'Seafood': '🐟', 'Rich Fish': '🐠',
+    'Shellfish': '🦐', 'Cheese': '🧀', 'Pasta': '🍝', 'Pizza': '🍕', 'Vegetables': '🥬', 'Vegetarian': '🥬',
+    'Veal': '🥩', 'Game Meat': '🦌', 'Barbecue': '🍖', 'Codfish': '🐟', 'Sweet Dessert': '🍰', 'Dessert': '🍰',
+    'Appetizers': '🍾', 'Fruit': '🍇', 'Nuts': '🥜', 'Chocolate': '🍫'
+}
+
+
+# --- OPTIMIZATION 1: Data Loading & Pre-processing ---
+@lru_cache(maxsize=1)
+def load_and_preprocess_data():
+    """Loads and performs expensive one-time preprocessing on the dataset."""
+    try:
+        df = pd.read_csv('XWines_Full_100K_wines.csv')
+    except FileNotFoundError:
+        raise FileNotFoundError("CSV file 'XWines_Full_100K_wines.csv' not found.")
+
+    def parse_list_string(s):
+        try:
+            return ast.literal_eval(s) if isinstance(s, str) else []
+        except (ValueError, SyntaxError):
+            return []
+
+    df['grapes_list'] = df['Grapes'].apply(parse_list_string)
+    df['harmonize_list'] = df['Harmonize'].apply(parse_list_string)
+    df['main_grape'] = df['grapes_list'].apply(lambda x: x[0] if x else 'Unknown')
+    df['num_grapes'] = df['grapes_list'].apply(len)
+    df['body_numeric'] = df['Body'].map(BODY_MAPPING)
+    return df
+
+
+# --- OPTIMIZATION 2: Vectorized Data Aggregation ---
+def get_top_food_pairings(series, top_n=3):
+    """Get top N food pairings with emojis and names."""
+    all_pairings = [item for sublist in series for item in sublist]
+    if not all_pairings:
+        return {'emojis': '🍽️', 'names': 'General'}
+    
+    top_items = Counter(all_pairings).most_common(top_n)
+    emojis = ''.join([FOOD_EMOJIS.get(item[0], '🍽️') for item in top_items])
+    names = ', '.join([item[0] for item in top_items])
+    return {'emojis': emojis, 'names': names}
+
+
+def aggregate_wine_data(df, wine_types, max_grape_count, min_samples_choice, regional_grouping):
+    """Filters and aggregates wine data using efficient, vectorized pandas operations."""
+    filtered_df = df.copy()
+
+    if wine_types and 'All' not in wine_types:
+        filtered_df = filtered_df[filtered_df['Type'].isin(wine_types)]
+    if max_grape_count < 5:
+        filtered_df = filtered_df[filtered_df['num_grapes'] <= max_grape_count]
+
+    group_by_cols = ['main_grape', 'Type']
+    if regional_grouping:
+        group_by_cols.append('Country')
+
+    agg_df = filtered_df.groupby(group_by_cols).agg(
+        count=('ABV', 'size'),
+        avg_fullness=('body_numeric', 'mean'),
+        abv_list=('ABV', list),
+        body_list=('Body', list),
+        acidity_list=('Acidity', list),
+        harmonize_list=('harmonize_list', list),
+        region_count=('RegionName', 'nunique'),
+        winery_count=('WineryName', 'nunique')
+    ).reset_index()
+
+    min_samples = SAMPLE_THRESHOLDS[min_samples_choice]
+    agg_df = agg_df[agg_df['count'] >= min_samples].copy()
+
+    if agg_df.empty:
+        return agg_df
+
+    # --- THE FIX ---
+    agg_df['body_dist'] = agg_df['body_list'].apply(
+        lambda x: (pd.Series(x).value_counts(normalize=True) * 100).to_dict())
+    agg_df['acid_dist'] = agg_df['acidity_list'].apply(
+        lambda x: (pd.Series(x).value_counts(normalize=True) * 100).to_dict())
+    # --- END OF FIX ---
+
+    agg_df['pairing_data'] = agg_df['harmonize_list'].apply(get_top_food_pairings)
+    agg_df['pairing_emoji'] = agg_df['pairing_data'].apply(lambda x: x['emojis'])
+    agg_df['pairing_names'] = agg_df['pairing_data'].apply(lambda x: x['names'])
+    agg_df['wine_type_order'] = agg_df['Type'].map(WINE_TYPE_ORDER)
+    agg_df = agg_df.sort_values(by=['wine_type_order', 'avg_fullness'], ascending=[False, True])
+
+    return agg_df
+
+
+# --- OPTIMIZATION 3: Efficient & Clean Chart Creation ---
+def create_wine_chart(chart_data, regional_grouping):
+    """Creates the Plotly figure with optimized traces and layout."""
+    if chart_data.empty:
+        fig = go.Figure()
+        fig.add_annotation(text="No data available with current filters.", xref="paper", yref="paper", x=0.5, y=0.5,
+                           showarrow=False)
+        return fig
+
+    num_rows = len(chart_data)
+
+    # Add wine type emoji based on type
+    wine_type_emojis = {'Red': '🍷', 'White': '🥂', 'Rosé': '🌸', 'Sparkling': '🍾'}
+    chart_data['wine_emoji'] = chart_data['Type'].map(wine_type_emojis).fillna('🍷')
+    
+    if regional_grouping:
+        chart_data['flag'] = chart_data['Country'].map(COUNTRY_FLAGS).fillna('🌍')
+        chart_data['grape_label'] = chart_data.apply(lambda row: f"{row['wine_emoji']} {row['main_grape']} {row['flag']}", axis=1)
+    else:
+        chart_data['grape_label'] = chart_data.apply(lambda row: f"{row['wine_emoji']} {row['main_grape']}", axis=1)
+
+    y_labels = chart_data['grape_label'].tolist()
+
+    fig = make_subplots(
+        rows=1, cols=5,
+        specs=[[{}, {"type": "bar"}, {"type": "bar"}, {"type": "box"}, {}]],
+        column_widths=[0.30, 0.25, 0.25, 0.15, 0.05],
+        horizontal_spacing=0.02,
+        shared_yaxes=True
+    )
+
+    hover_texts = chart_data.apply(
+        lambda row: (
+            f"<b>{row['main_grape']} ({row.get('Country', 'Global') if regional_grouping else 'Global'})</b><br>"
+            f"Wineries: {row['winery_count']}<br>"
+            f"Regions: {row['region_count']}<br>"
+            f"Total Wines: {row['count']:,}"),
+        axis=1
+    )
+    fig.add_trace(go.Bar(
+        y=y_labels, x=[1] * num_rows, orientation='h',
+        marker_color='rgba(0,0,0,0)', showlegend=False,
+        hoverinfo='text', hovertext=hover_texts
+    ), row=1, col=1)
+    fig.add_trace(go.Scatter(
+        y=y_labels, x=[0.03] * num_rows, mode='text',
+        text=y_labels, textposition='middle right',
+        textfont={'size': 14, 'color': '#2F2F2F'},
+        hoverinfo='none', showlegend=False
+    ), row=1, col=1)
+
+    body_colors = {'Very light-bodied': '#FFB6C1', 'Light-bodied': '#CD5C5C', 'Medium-bodied': '#C13636',
+                   'Full-bodied': '#8B0000', 'Very full-bodied': '#4B0000'}
+    for body_type in BODY_ORDER:
+        values = chart_data['body_dist'].apply(lambda d: d.get(body_type, 0))
+        fig.add_trace(go.Bar(
+            y=y_labels, x=values, name=body_type, orientation='h',
+            marker_color=body_colors.get(body_type), showlegend=False,
+            hovertemplate=f"{body_type}: %{{x:.1f}}%<extra></extra>"
+        ), row=1, col=2)
+
+    acid_colors = {'Low': '#F5F5DC', 'Medium': '#DAA520', 'High': '#B8860B'}
+    for acid_type in ACIDITY_ORDER:
+        values = chart_data['acid_dist'].apply(lambda d: d.get(acid_type, 0))
+        fig.add_trace(go.Bar(
+            y=y_labels, x=values, name=acid_type, orientation='h',
+            marker_color=acid_colors.get(acid_type), showlegend=False,
+            hovertemplate=f"{acid_type} acidity: %{{x:.1f}}%<extra></extra>"
+        ), row=1, col=3)
+
+    # Color box plots by wine type
+    box_colors = {'Red': '#8B0000', 'White': '#DAA520', 'Rosé': '#CD5C5C', 'Sparkling': '#9370DB'}
+    for idx, (i, row) in enumerate(chart_data.iterrows()):
+        abv_values = row['abv_list']
+        color = box_colors.get(row['Type'], '#6A5ACD')
+        fig.add_trace(go.Box(
+            y=[y_labels[idx]] * len(abv_values), x=abv_values, name=row['Type'], orientation='h',
+            showlegend=False, marker_color=color, line_color=color,
+            hovertemplate=f"ABV: %{{x:.1f}}%<extra></extra>"
+        ), row=1, col=4)
+
+    fig.add_trace(go.Scatter(
+        y=y_labels, x=[0.5] * num_rows, mode='text',
+        text=chart_data['pairing_emoji'], textposition='middle center',
+        textfont={'size': 22}, showlegend=False,
+        hoverinfo='text', hovertext=chart_data['pairing_names']
+    ), row=1, col=5)
+
+    fig.update_layout(
+        title={
+            'text': "Wine Characteristics by Grape Variety",
+            'x': 0.5,
+            'font': {'size': 24, 'color': '#2F2F2F'}
+        },
+        height=max(500, num_rows * 40),
+        barmode='stack', showlegend=False, plot_bgcolor='#FFFFFF', paper_bgcolor='#F5F5F5',
+        margin=dict(l=10, r=10, t=120, b=20), boxgap=0.5, bargap=0.4
+    )
+
+    column_titles = ["Wine / Hover for Info", "Body Profile (%)", "Acidity Profile (%)", "Alcohol (ABV %)", "Food Pairing"]
+    for i, title in enumerate(column_titles, 1):
+        domain = fig.layout[f'xaxis{i if i > 1 else ""}'].domain
+        fig.add_annotation(
+            x=(domain[0] + domain[1]) / 2, y=1.05,
+            xref="paper", yref="paper", text=f"<b>{title}</b>",
+            xanchor='center', showarrow=False, font={'size': 14, 'color': '#2F2F2F'}
+        )
+
+    for i in range(1, 6):
+        fig.update_yaxes(showticklabels=False, showgrid=False, zeroline=False, row=1, col=i)
+        fig.update_xaxes(showticklabels=False, showgrid=False, zeroline=False, title_text="", row=1, col=i)
+
+    fig.update_yaxes(categoryorder="array", categoryarray=y_labels, autorange=False, range=[-0.5, num_rows - 0.5],
+                     row=1, col=1)
+
+    for i in range(num_rows):
+        if i % 2 == 1:
+            fig.add_hrect(y0=i - 0.5, y1=i + 0.5, fillcolor="#F0F0F0", layer="below", line_width=0, row=1, col="all")
+
+    return fig
+
+
+# --- Gradio Interface Logic ---
+def update_dashboard(wine_types, max_grape_count, min_samples_choice, regional_grouping,
+                     progress=gr.Progress(track_tqdm=True)):
+    """Main function to update dashboard."""
+    progress(0, desc="Loading and processing data...")
+    df = load_and_preprocess_data()
+
+    progress(0.5, desc="Filtering and aggregating...")
+    chart_data = aggregate_wine_data(df, wine_types, max_grape_count, min_samples_choice, regional_grouping)
+
+    progress(0.8, desc="Creating chart...")
+    fig = create_wine_chart(chart_data, regional_grouping)
+
+    total_combinations = len(chart_data)
+    total_wines = chart_data['count'].sum() if not chart_data.empty else 0
+    min_samples = SAMPLE_THRESHOLDS[min_samples_choice]
+    grouping_type = "grape+region" if regional_grouping else "grape+type"
+    summary = f"📊 Showing **{total_combinations}** {grouping_type} combinations from **{total_wines:,}** wines (min {min_samples} samples each)"
+
+    return fig, summary
+
+
+# Create Gradio interface
+def create_interface():
+    with gr.Blocks(title="Wine Analysis Dashboard", theme=gr.themes.Soft()) as demo:
+        gr.Markdown("# 🍷 Wine Characteristics Dashboard")
+
+        with gr.Row():
+            wine_type_filter = gr.CheckboxGroup(
+                choices=['Red', 'White', 'Rosé', 'Sparkling', 'Dessert', 'Dessert/Port'],
+                value=['Red'], label="🍷 Wine Types"
+            )
+            max_grape_slider = gr.Slider(
+                minimum=1, maximum=5, step=1, value=1, label="🍇 Max Grapes per Wine",
+                info="1: Varietals, 5: All Blends"
+            )
+            min_samples_choice = gr.Radio(
+                choices=list(SAMPLE_THRESHOLDS.keys()), value='Common (100+)',
+                label="Minimum Sample Size"
+            )
+            regional_grouping = gr.Checkbox(
+                value=True, label="Split By Country"
+            )
+
+        summary_text = gr.Markdown()
+        wine_plot = gr.Plot()
+
+        inputs = [wine_type_filter, max_grape_slider, min_samples_choice, regional_grouping]
+        outputs = [wine_plot, summary_text]
+
+        # Auto-update when any input changes
+        for input_component in inputs:
+            input_component.change(update_dashboard, inputs=inputs, outputs=outputs)
+        
+        demo.load(fn=update_dashboard, inputs=inputs, outputs=outputs)
+
+    return demo
+
+
+if __name__ == "__main__":
+    app_interface = create_interface()
+    app_interface.launch()