d4c-datapkg-statistical/experiments/boundaries_spatial_checks.ipynb

{
 "cells": [
  {
   "cell_type": "code",
   "execution_count": 30,
   "id": "3849cfa2-bb10-4323-95a3-51d205f497d5",
   "metadata": {},
   "outputs": [
    {
     "name": "stdout",
     "output_type": "stream",
     "text": [
      "Buckaroo has been enabled as the default DataFrame viewer.  To return to default dataframe visualization use `from buckaroo import disable; disable()`\n"
     ]
    }
   ],
   "source": [
    "import os\n",
    "\n",
    "import buckaroo\n",
    "import geopandas as gpd\n",
    "from IPython.core.interactiveshell import InteractiveShell\n",
    "from lonboard import viz\n",
    "import pandas as pd\n",
    "from sqlalchemy import create_engine\n",
    "\n",
    "# Enable multiple outputs per cell\n",
    "InteractiveShell.ast_node_interactivity = \"all\"\n",
    "# Show all columns\n",
    "pd.set_option('display.max_columns', None)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 31,
   "id": "7e8b2e03-594c-4d8d-8914-ff4fe68117ac",
   "metadata": {},
   "outputs": [],
   "source": [
    "DATABASE = os.getenv('POSTGRES_DB')\n",
    "USER = os.getenv('POSTGRES_USER')\n",
    "PASSWORD = os.getenv('POSTGRES_PASSWORD')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 32,
   "id": "dca70ec0-35c5-4c51-9e6a-5131ed46522c",
   "metadata": {},
   "outputs": [],
   "source": [
    "engine = create_engine(f'postgresql://{USER}:{PASSWORD}@db:5432/{DATABASE}')"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 34,
   "id": "08935c71-ff7b-4a2a-aa9d-029b4231cc71",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "95c1fb1edb9e440d916fa8cd3496bb46",
       "version_major": 2,
       "version_minor": 1
      },
      "text/plain": [
       "GeopandasBuckarooWidget(buckaroo_options={'sampled': ['random'], 'auto_clean': ['aggressive', 'conservative'],…"
      ]
     },
     "metadata": {},
     "output_type": "display_data"
    }
   ],
   "source": [
    "sql = \"\"\"\n",
    "SELECT * FROM silver.pr_2016;\n",
    "\"\"\"\n",
    "df = gpd.read_postgis(sql, con=engine)\n",
    "df"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": 27,
   "id": "03ff2898-1318-4859-8249-d2c3a05117a7",
   "metadata": {},
   "outputs": [
    {
     "data": {
      "application/vnd.jupyter.widget-view+json": {
       "model_id": "775652f4d632485abaf8babe7c671b25",
       "version_major": 2,
       "version_minor": 1
      },
      "text/plain": [
       "Map(basemap_style=<CartoBasemap.DarkMatter: 'https://basemaps.cartocdn.com/gl/dark-matter-gl-style/style.json'…"
      ]
     },
     "execution_count": 27,
     "metadata": {},
     "output_type": "execute_result"
    }
   ],
   "source": [
    "viz(df)"
   ]
  },
  {
   "cell_type": "code",
   "execution_count": null,
   "id": "bfcbf94c-0d65-422d-90b3-2a8d79ef1c32",
   "metadata": {},
   "outputs": [],
   "source": []
  }
 ],
 "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": "3.12.3"
  }
 },
 "nbformat": 4,
 "nbformat_minor": 5
}