hexagon_bokeh (3).ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 1,
   "id": "2b2c5aac-44e1-4a55-9ca1-f99c45d9df4a",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "/Users/shatakshibhatnagar/Downloads/HexagonDSRS\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "print(os.getcwd())  # Prints the current working directory\n",
    "import pandas as pd # For data manipulation and reading Excel files\n",
    "\n",
    "# Step 1: Load datasets\n",
    "# Load synthetic dataset with state and subcategory data\n",
    "\n",
    "path = '/Users/shatakshibhatnagar/Downloads/HexagonDSRS/synthetic_data (7) (1) (1).xlsx'\n",
    "synthetic_df = pd.read_excel(path)\n",
    "\n",
    "# Load hexagonal map data containing state coordinates and abbreviations\n",
    "\n",
    "hex_map_data_path = '/Users/shatakshibhatnagar/Downloads/HexagonDSRS/Polygonic Hex Map New.xlsx'\n",
    "hex_df = pd.read_excel(hex_map_data_path, sheet_name='Sheet1')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 5,
   "id": "8b1de674-3795-4d53-b9ab-a0ca21bbe0b2",
   "metadata": {},
   "outputs": [],
   "source": [
    "# Import Bokeh libraries for plotting\n",
    "from bokeh.plotting import figure, curdoc\n",
    "from bokeh.models import ColumnDataSource, TapTool\n",
    "from bokeh.layouts import row\n",
    "from bokeh.palettes import Category20c\n",
    "from bokeh.transform import cumsum\n",
    "import math\n",
    "from math import pi, cos, sin\n",
    "import pandas as pd\n",
    "from bokeh.palettes import Viridis256\n",
    "from bokeh.palettes import Pastel1\n",
    "\n",
    "\n",
    "# Step 2: Aggregate data for pie chart\n",
    "# Group by 'state' and 'subcategory' to calculate counts\n",
    "agg_data = synthetic_df.groupby(['state', 'subcategory']).size().reset_index(name='Count')\n",
    "\n",
    "# Add a column for percentages\n",
    "agg_data['percentage'] = (agg_data['Count'] / agg_data.groupby('state')['Count'].transform('sum')) * 100\n",
    "\n",
    "# Step 3: Generate Hexagon Coordinates\n",
    "# Calculate centroids for hexagons based on 'State' columns in the hex_df dataset\n",
    "centroids = hex_df.groupby('State')[['X', 'Y']].mean().to_dict(orient='index')\n",
    "\n",
    "# Prepare data structure for hexagon plotting\n",
    "hex_data = {\n",
    "    'x': [],\n",
    "    'y': [],\n",
    "    'state': [],\n",
    "    'abbreviation': [],\n",
    "    'color': [],\n",
    "    'centroid_x': [],  # Add centroid X coordinates\n",
    "    'centroid_y': [] \n",
    "}\n",
    "\n",
    "# Assign unique colors to each state\n",
    "unique_states = hex_df['State'].unique()\n",
    "hex_colors = Category20c[len(unique_states)] if len(unique_states) <= 20 else Category20c[20]\n",
    "state_colors = {state: hex_colors[i % len(hex_colors)] for i, state in enumerate(unique_states)}\n",
    "\n",
    "# Generate coordinates for each state\n",
    "for state, coords in centroids.items():\n",
    "    hex_coords = [(coords['X'] + 0.8 * math.cos(2 * pi * i / 6), coords['Y'] + 0.8 * math.sin(2 * pi * i / 6)) for i in range(6)]\n",
    "    hex_x, hex_y = zip(*hex_coords)\n",
    "    hex_data['x'].append(list(hex_x) + [hex_x[0]])\n",
    "    hex_data['y'].append(list(hex_y) + [hex_y[0]])\n",
    "    hex_data['state'].append(state)\n",
    "    if state in hex_df['State'].values:\n",
    "        hex_data['abbreviation'].append(hex_df[hex_df['State'] == state]['Abbreviation'].iloc[0])\n",
    "    else:\n",
    "        hex_data['abbreviation'].append('N/A')\n",
    "    hex_data['color'].append(state_colors[state])\n",
    "    hex_data['centroid_x'].append(coords['X'])  # Use X centroid for labels\n",
    "    hex_data['centroid_y'].append(coords['Y'])  # Use Y centroid for labels\n",
    "\n",
    "# Convert hexagon data to Bokeh's ColumnDataSource for plotting\n",
    "hex_source = ColumnDataSource(hex_data)\n",
    "\n",
    "# Step 4: Initialize Pie Chart Data\n",
    "# Create an empty data source for the pie chart, which will update dynamically\n",
    "pie_source = ColumnDataSource(data=dict(subcategory=[], angle=[], pie_color=[], percentage_label=[], x_label=[], y_label=[]))\n",
    "\n",
    "# Step 5: Create Hexagon Plot\n",
    "hex_plot = figure(title=\"Visualizing Regional Product Adoptio\", tools=\"tap\", tooltips=\"@state\", width=1200, height=900)\n",
    "hex_plot.add_tools(TapTool())\n",
    "hex_plot.patches('x', 'y', source=hex_source, fill_alpha=0.8, fill_color='color', line_color=\"white\")\n",
    "\n",
    "# Add state abbreviations at centroids\n",
    "hex_plot.text(\n",
    "    x='centroid_x', y='centroid_y', text='abbreviation',\n",
    "    source=hex_source, text_align=\"center\", text_baseline=\"middle\",\n",
    "    text_color=\"black\", text_font_size=\"10pt\"\n",
    ")\n",
    "\n",
    "# Hide axes and grid for a cleaner look\n",
    "hex_plot.axis.visible = False\n",
    "hex_plot.grid.grid_line_color = None\n",
    "\n",
    "# Step 6: Create Pie Chart\n",
    "from bokeh.models import LabelSet\n",
    "pie_chart = figure(title=\"Pie Chart\", width=400, height=400, tools=\"\")\n",
    "pie_chart.wedge(\n",
    "    x=0, y=0, radius=0.8,\n",
    "    start_angle='start_angle',\n",
    "    end_angle='end_angle',\n",
    "    line_color=\"white\",\n",
    "    fill_color='pie_color',\n",
    "    source=pie_source,\n",
    "    legend_field='subcategory'\n",
    ")\n",
    "\n",
    "pie_chart.axis.visible = False\n",
    "pie_chart.grid.grid_line_color = None\n",
    "\n",
    "# Add labels to pie chart\n",
    "labels = LabelSet(\n",
    "    x='x_label', y='y_label', text='percentage_label',\n",
    "    source=pie_source,\n",
    "    text_align='center', text_baseline='middle',\n",
    "    text_font_size=\"10pt\", text_color=\"black\"\n",
    ")\n",
    "\n",
    "pie_chart.add_layout(labels)\n",
    "\n",
    "# Callback function to update the pie chart based on hexagon selection\n",
    "def update_pie_chart(attr, old, new):\n",
    "    selected = hex_source.selected.indices\n",
    "    if not selected:\n",
    "        # Clear pie chart if no hexagon is selected\n",
    "        pie_source.data = dict(\n",
    "            subcategory=[], start_angle=[], end_angle=[], pie_color=[], percentage_label=[], x_label=[], y_label=[]\n",
    "        )\n",
    "        pie_chart.title.text = \"No State Selected\"\n",
    "        return\n",
    "\n",
    "    selected_state = hex_source.data['state'][selected[0]]\n",
    "    print(f\"Selected State: {selected_state}\")\n",
    "    state_data = agg_data[agg_data['state'] == selected_state]\n",
    "\n",
    "    if state_data.empty:\n",
    "        # Handle missing data for the selected state\n",
    "        print(f\"No subcategories found for {selected_state}.\")\n",
    "        pie_source.data = dict(\n",
    "            subcategory=[], start_angle=[], end_angle=[], pie_color=[], percentage_label=[], x_label=[], y_label=[]\n",
    "        )\n",
    "        pie_chart.title.text = f\"No Data Available for {selected_state.capitalize()}\"\n",
    "        return\n",
    "\n",
    "    # Calculate angles and cumulative sum for the pie chart\n",
    "    state_data = state_data.copy()\n",
    "    state_data['angle'] = state_data['percentage'] / 100 * 2 * pi\n",
    "    state_data['start_angle'] = state_data['angle'].cumsum().shift(fill_value=0)\n",
    "    state_data['end_angle'] = state_data['start_angle'] + state_data['angle']\n",
    "    state_data['percentage_label'] = state_data['percentage'].round(1).astype(str) + '%'  # Format percentages\n",
    "\n",
    "\n",
    "    # Dynamically assign lighter colors to each subcategory\n",
    "    num_categories = len(state_data)\n",
    "    palette = list(Pastel1[8])  # Use Pastel1 for lighter colors, max size is 8\n",
    "    if num_categories > len(palette):  # Extend the palette if needed\n",
    "        palette = (palette * (num_categories // len(palette) + 1))[:num_categories]\n",
    "    state_data['pie_color'] = palette[:num_categories] \n",
    "\n",
    "    # Compute label positions\n",
    "    state_data['x_label'] = [\n",
    "        0.4 * math.cos(state_data['start_angle'].iloc[i] + state_data['angle'].iloc[i] / 2)\n",
    "        for i in range(len(state_data))\n",
    "    ]\n",
    "    state_data['y_label'] = [\n",
    "        0.4 * math.sin(state_data['start_angle'].iloc[i] + state_data['angle'].iloc[i] / 2)\n",
    "        for i in range(len(state_data))\n",
    "    ]\n",
    "\n",
    "    # Convert pandas Series to lists for Bokeh compatibility\n",
    "    pie_source.data = {\n",
    "        'subcategory': state_data['subcategory'].tolist(),\n",
    "        'start_angle': state_data['start_angle'].tolist(),\n",
    "        'end_angle': state_data['end_angle'].tolist(),\n",
    "        'pie_color': state_data['pie_color'].tolist(),  # Explicitly convert to list\n",
    "        'percentage_label': state_data['percentage_label'].tolist(),\n",
    "        'x_label': state_data['x_label'],  # Already computed as lists\n",
    "        'y_label': state_data['y_label'],  # Already computed as lists\n",
    "    }\n",
    "    pie_chart.title.text = f\"Subcategory Distribution for {selected_state.capitalize()}\"\n",
    "    print(f\"Pie Source Data: {pie_source.data}\")  # Debugging line\n",
    "\n",
    "\n",
    "# Attach the callback\n",
    "hex_source.selected.on_change('indices', update_pie_chart)\n",
    " \n",
    "# Step 7: Layout and Show\n",
    "curdoc().add_root(row(hex_plot, pie_chart))\n"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "f7c6eea9-1526-4a3c-bc25-8fc33fc2b95d",
   "metadata": {},
   "outputs": [],
   "source": [
    "#Steps to Run the Code - \n",
    "#Save this script as hexagon_plot.py.\n",
    "#Open the Command Prompt or terminal\n",
    "#Navigate to the script's directory using:\n",
    "#Copy code - cd C:\\path\\to\\script\n",
    "#Run the script with:\n",
    "#Copy code - bokeh serve --show hexagon_plot.py"
   ]
  }
 ],
 "metadata": {
  "kernelspec": {
   "display_name": "Python 3 (ipykernel)",
   "language": "python",
   "name": "python3"
  },
  "language_info": {
   "codemirror_mode": {
    "name": "ipython",
    "version": 3
   },
   "file_extension": ".py",
   "mimetype": "text/x-python",
   "name": "python",
   "nbconvert_exporter": "python",
   "pygments_lexer": "ipython3",
   "version": "3.12.4"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}