From 9e828e70580386119943cab25db488a2e26929a0 Mon Sep 17 00:00:00 2001
From: Chandrasekhar Ramakrishnan <cramakri@ethz.ch>
Date: Thu, 2 Apr 2020 20:28:17 +0000
Subject: [PATCH] feat: implement notebook for getting population data from
Wikidata
---
notebooks/process/wikidata-pop-data.ipynb | 136 ++++++++++++++++++++++
1 file changed, 136 insertions(+)
create mode 100644 notebooks/process/wikidata-pop-data.ipynb
diff --git a/notebooks/process/wikidata-pop-data.ipynb b/notebooks/process/wikidata-pop-data.ipynb
new file mode 100644
index 000000000..1eefbd38d
--- /dev/null
+++ b/notebooks/process/wikidata-pop-data.ipynb
@@ -0,0 +1,136 @@
+{
+ "cells": [
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "# Gather Population Data from Wikidata"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "import os\n",
+ "import pandas as pd\n",
+ "\n",
+ "from covid_19_dashboard import helper"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {
+ "tags": [
+ "parameters"
+ ]
+ },
+ "outputs": [],
+ "source": [
+ "out_folder = '../../data/atlas/wikidata'\n",
+ "PAPERMILL_OUTPUT_PATH = None"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "def write_population_data(iso_code, df):\n",
+ " out_path = os.path.join(out_folder, f\"{iso_code.lower()}-population.csv\")\n",
+ " print(f\"Writing {len(df)} rows to {out_path}\")\n",
+ " if PAPERMILL_OUTPUT_PATH is None:\n",
+ " return\n",
+ " df.to_csv(out_path)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Italy"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "iso_code = \"ITA\"\n",
+ "pops = helper.get_region_populations(\n",
+ " iso_code,\n",
+ " additional_fields=\"?istatid\",\n",
+ " additional_query=\"?region wdt:P635 ?istatid .\",\n",
+ ")\n",
+ "df = pd.DataFrame(pops)\n",
+ "write_population_data(iso_code, df)\n",
+ "df.head(2)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## Switzerland"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "iso_code = \"CHE\"\n",
+ "pops = helper.get_region_populations(iso_code)\n",
+ "df = pd.DataFrame(pops)\n",
+ "write_population_data(iso_code, df)\n",
+ "df.head(2)"
+ ]
+ },
+ {
+ "cell_type": "markdown",
+ "metadata": {},
+ "source": [
+ "## United States"
+ ]
+ },
+ {
+ "cell_type": "code",
+ "execution_count": null,
+ "metadata": {},
+ "outputs": [],
+ "source": [
+ "iso_code = \"USA\"\n",
+ "pops = helper.get_region_populations(iso_code)\n",
+ "df = pd.DataFrame(pops)\n",
+ "write_population_data(iso_code, df)\n",
+ "df.head(2)"
+ ]
+ }
+ ],
+ "metadata": {
+ "kernelspec": {
+ "display_name": "Python 3",
+ "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": "3.7.3"
+ }
+ },
+ "nbformat": 4,
+ "nbformat_minor": 4
+}
--
GitLab