{
"cells": [
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"# Employment rates in Finland"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"**Goal:** plot an interactive map of employment rates across Finnish regions.\n",
"\n",
"**Required modules:**\n",
"- Folium for plotting interactive maps based on leaflet.js\n",
"- Pandas for handling tabular data\n",
"- Geopandas for handling spatial data"
]
},
{
"cell_type": "code",
"execution_count": 1,
"metadata": {
"deletable": true,
"editable": true
},
"outputs": [],
"source": [
"# Import required modules:\n",
"import folium\n",
"import pandas as pd\n",
"import geopandas as gpd\n",
"import matplotlib.pyplot as plt\n",
"%matplotlib inline"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"### Employment rate data \n",
"\n",
"Employment rate refers to \"the proportion of the employed among persons aged 15 to 64\". Data from Statistics Finland (saved from .xml-file to csv in Excel..)."
]
},
{
"cell_type": "code",
"execution_count": 2,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true,
"jupyter": {
"outputs_hidden": false
}
},
"outputs": [
{
"data": {
"text/html": [
"
\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" seutukunta | \n",
" seutukunta_nimi | \n",
" tyollisyys | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" SK011 | \n",
" Helsingin seutukunta | \n",
" 73.0 | \n",
"
\n",
" \n",
" 1 | \n",
" SK014 | \n",
" Raaseporin seutukunta | \n",
" 70.3 | \n",
"
\n",
" \n",
" 2 | \n",
" SK015 | \n",
" Porvoon seutukunta | \n",
" 74.3 | \n",
"
\n",
" \n",
" 3 | \n",
" SK016 | \n",
" Loviisan seutukunta | \n",
" 71.5 | \n",
"
\n",
" \n",
" 4 | \n",
" SK021 | \n",
" Åboland-Turunmaan seutukunta | \n",
" 72.9 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" seutukunta seutukunta_nimi tyollisyys\n",
"0 SK011 Helsingin seutukunta 73.0\n",
"1 SK014 Raaseporin seutukunta 70.3\n",
"2 SK015 Porvoon seutukunta 74.3\n",
"3 SK016 Loviisan seutukunta 71.5\n",
"4 SK021 Åboland-Turunmaan seutukunta 72.9"
]
},
"execution_count": 2,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"# Read in data\n",
"data = pd.read_csv(\"data/seutukunta_tyollisyys_2013.csv\", sep=\",\")\n",
"data.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"### Sub-regional units\n",
"\n",
"The spatial data for the [sub-regional units](http://www.stat.fi/meta/luokitukset/seutukunta/001-2018/index_en.html#_ga=2.189315043.975495398.1543480533-446957274.1543480533) (*Seutukunnat* in Finnish) can be retrieved from the Statistics Finland Web Feature Service `http://geo.stat.fi/geoserver/tilastointialueet/wfs`"
]
},
{
"cell_type": "code",
"execution_count": 3,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true,
"jupyter": {
"outputs_hidden": false
}
},
"outputs": [],
"source": [
"# A layer saved to GeoJson in QGIS..\n",
"#geodata = gpd.read_file('Seutukunnat_2018.geojson')\n",
"\n",
"# Get features directly from the wfs\n",
"url = \"http://geo.stat.fi/geoserver/tilastointialueet/wfs?request=GetFeature&typename=tilastointialueet:seutukunta1000k_2018&outputformat=JSON\"\n",
"geodata = gpd.read_file(url)"
]
},
{
"cell_type": "code",
"execution_count": 4,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true,
"jupyter": {
"outputs_hidden": false
}
},
"outputs": [
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" id | \n",
" vuosi | \n",
" seutukunta | \n",
" nimi | \n",
" namn | \n",
" name | \n",
" geometry | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" seutukunta1000k_2018.1 | \n",
" 2018 | \n",
" 011 | \n",
" Helsinki | \n",
" Helsingfors | \n",
" Helsinki | \n",
" MULTIPOLYGON (((409963.522 6681658.341, 409969... | \n",
"
\n",
" \n",
" 1 | \n",
" seutukunta1000k_2018.2 | \n",
" 2018 | \n",
" 014 | \n",
" Raasepori | \n",
" Raseborg | \n",
" Raasepori | \n",
" MULTIPOLYGON (((306616.919 6665438.489, 306668... | \n",
"
\n",
" \n",
" 2 | \n",
" seutukunta1000k_2018.3 | \n",
" 2018 | \n",
" 015 | \n",
" Porvoo | \n",
" Borgå | \n",
" Porvoo | \n",
" MULTIPOLYGON (((427108.141 6694151.025, 427175... | \n",
"
\n",
" \n",
" 3 | \n",
" seutukunta1000k_2018.4 | \n",
" 2018 | \n",
" 016 | \n",
" Loviisa | \n",
" Lovisa | \n",
" Loviisa | \n",
" MULTIPOLYGON (((444038.768 6703649.355, 444155... | \n",
"
\n",
" \n",
" 4 | \n",
" seutukunta1000k_2018.5 | \n",
" 2018 | \n",
" 021 | \n",
" Åboland-Turunmaa | \n",
" Åboland-Turunmaa | \n",
" Åboland-Turunmaa | \n",
" MULTIPOLYGON (((190999.717 6715878.622, 191021... | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" id vuosi seutukunta \\\n",
"0 seutukunta1000k_2018.1 2018 011 \n",
"1 seutukunta1000k_2018.2 2018 014 \n",
"2 seutukunta1000k_2018.3 2018 015 \n",
"3 seutukunta1000k_2018.4 2018 016 \n",
"4 seutukunta1000k_2018.5 2018 021 \n",
"\n",
" nimi \\\n",
"0 Helsinki \n",
"1 Raasepori \n",
"2 Porvoo \n",
"3 Loviisa \n",
"4 Åboland-Turunmaa \n",
"\n",
" namn \\\n",
"0 Helsingfors \n",
"1 Raseborg \n",
"2 Borgå \n",
"3 Lovisa \n",
"4 Åboland-Turunmaa \n",
"\n",
" name \\\n",
"0 Helsinki \n",
"1 Raasepori \n",
"2 Porvoo \n",
"3 Loviisa \n",
"4 Åboland-Turunmaa \n",
"\n",
" geometry \n",
"0 MULTIPOLYGON (((409963.522 6681658.341, 409969... \n",
"1 MULTIPOLYGON (((306616.919 6665438.489, 306668... \n",
"2 MULTIPOLYGON (((427108.141 6694151.025, 427175... \n",
"3 MULTIPOLYGON (((444038.768 6703649.355, 444155... \n",
"4 MULTIPOLYGON (((190999.717 6715878.622, 191021... "
]
},
"execution_count": 4,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"geodata.head()"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"### Join attributes and geometries\n",
"\n",
"We can join the attribute layer and spatial layer based on the region code (stored in column 'seutukunta'). The region codes in the csv contain additional letters \"SK\" which we need to remove before the join:"
]
},
{
"cell_type": "code",
"execution_count": 5,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true,
"jupyter": {
"outputs_hidden": false
}
},
"outputs": [
{
"data": {
"text/plain": [
"0 011\n",
"1 014\n",
"2 015\n",
"3 016\n",
"4 021\n",
"Name: seutukunta, dtype: object"
]
},
"execution_count": 5,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"data[\"seutukunta\"] = data[\"seutukunta\"].apply(lambda x: x[2:])\n",
"data[\"seutukunta\"].head()"
]
},
{
"cell_type": "markdown",
"metadata": {},
"source": [
"Now we can join the data based on the \"seutukunta\" -column.\n",
"Let's also check that we have a matching number of records before and after the join:"
]
},
{
"cell_type": "code",
"execution_count": 6,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true,
"jupyter": {
"outputs_hidden": false
}
},
"outputs": [
{
"name": "stdout",
"output_type": "stream",
"text": [
"Count of original attributes: 70\n",
"Count of original geometries: 70\n",
"Count after the join: 70\n"
]
},
{
"data": {
"text/html": [
"\n",
"\n",
"
\n",
" \n",
" \n",
" | \n",
" id | \n",
" vuosi | \n",
" seutukunta | \n",
" nimi | \n",
" namn | \n",
" name | \n",
" geometry | \n",
" seutukunta_nimi | \n",
" tyollisyys | \n",
"
\n",
" \n",
" \n",
" \n",
" 0 | \n",
" seutukunta1000k_2018.1 | \n",
" 2018 | \n",
" 011 | \n",
" Helsinki | \n",
" Helsingfors | \n",
" Helsinki | \n",
" MULTIPOLYGON (((409963.522 6681658.341, 409969... | \n",
" Helsingin seutukunta | \n",
" 73.0 | \n",
"
\n",
" \n",
" 1 | \n",
" seutukunta1000k_2018.2 | \n",
" 2018 | \n",
" 014 | \n",
" Raasepori | \n",
" Raseborg | \n",
" Raasepori | \n",
" MULTIPOLYGON (((306616.919 6665438.489, 306668... | \n",
" Raaseporin seutukunta | \n",
" 70.3 | \n",
"
\n",
" \n",
" 2 | \n",
" seutukunta1000k_2018.3 | \n",
" 2018 | \n",
" 015 | \n",
" Porvoo | \n",
" Borgå | \n",
" Porvoo | \n",
" MULTIPOLYGON (((427108.141 6694151.025, 427175... | \n",
" Porvoon seutukunta | \n",
" 74.3 | \n",
"
\n",
" \n",
" 3 | \n",
" seutukunta1000k_2018.4 | \n",
" 2018 | \n",
" 016 | \n",
" Loviisa | \n",
" Lovisa | \n",
" Loviisa | \n",
" MULTIPOLYGON (((444038.768 6703649.355, 444155... | \n",
" Loviisan seutukunta | \n",
" 71.5 | \n",
"
\n",
" \n",
" 4 | \n",
" seutukunta1000k_2018.5 | \n",
" 2018 | \n",
" 021 | \n",
" Åboland-Turunmaa | \n",
" Åboland-Turunmaa | \n",
" Åboland-Turunmaa | \n",
" MULTIPOLYGON (((190999.717 6715878.622, 191021... | \n",
" Åboland-Turunmaan seutukunta | \n",
" 72.9 | \n",
"
\n",
" \n",
"
\n",
"
"
],
"text/plain": [
" id vuosi seutukunta \\\n",
"0 seutukunta1000k_2018.1 2018 011 \n",
"1 seutukunta1000k_2018.2 2018 014 \n",
"2 seutukunta1000k_2018.3 2018 015 \n",
"3 seutukunta1000k_2018.4 2018 016 \n",
"4 seutukunta1000k_2018.5 2018 021 \n",
"\n",
" nimi \\\n",
"0 Helsinki \n",
"1 Raasepori \n",
"2 Porvoo \n",
"3 Loviisa \n",
"4 Åboland-Turunmaa \n",
"\n",
" namn \\\n",
"0 Helsingfors \n",
"1 Raseborg \n",
"2 Borgå \n",
"3 Lovisa \n",
"4 Åboland-Turunmaa \n",
"\n",
" name \\\n",
"0 Helsinki \n",
"1 Raasepori \n",
"2 Porvoo \n",
"3 Loviisa \n",
"4 Åboland-Turunmaa \n",
"\n",
" geometry \\\n",
"0 MULTIPOLYGON (((409963.522 6681658.341, 409969... \n",
"1 MULTIPOLYGON (((306616.919 6665438.489, 306668... \n",
"2 MULTIPOLYGON (((427108.141 6694151.025, 427175... \n",
"3 MULTIPOLYGON (((444038.768 6703649.355, 444155... \n",
"4 MULTIPOLYGON (((190999.717 6715878.622, 191021... \n",
"\n",
" seutukunta_nimi tyollisyys \n",
"0 Helsingin seutukunta 73.0 \n",
"1 Raaseporin seutukunta 70.3 \n",
"2 Porvoon seutukunta 74.3 \n",
"3 Loviisan seutukunta 71.5 \n",
"4 Åboland-Turunmaan seutukunta 72.9 "
]
},
"execution_count": 6,
"metadata": {},
"output_type": "execute_result"
}
],
"source": [
"#print info\n",
"print(\"Count of original attributes:\", len(data))\n",
"print(\"Count of original geometries:\", len(geodata))\n",
"\n",
"# Merge data\n",
"geodata = geodata.merge(data, on = \"seutukunta\")\n",
"\n",
"#Print info\n",
"print(\"Count after the join:\", len(geodata))\n",
"\n",
"geodata.head()"
]
},
{
"cell_type": "code",
"execution_count": 7,
"metadata": {},
"outputs": [],
"source": [
"## Create a static map"
]
},
{
"cell_type": "markdown",
"metadata": {
"deletable": true,
"editable": true
},
"source": [
"Now we have a spatial layer with the employment rate information (in column \"tyollisuus\").\n",
"Let's create a simple plot based on this data:"
]
},
{
"cell_type": "code",
"execution_count": 8,
"metadata": {
"collapsed": false,
"deletable": true,
"editable": true,
"jupyter": {
"outputs_hidden": false
}
},
"outputs": [
{
"data": {
"text/plain": [
""
]
},
"execution_count": 8,
"metadata": {},
"output_type": "execute_result"
},
{
"data": {
"image/png": "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\n",
"text/plain": [
"