diff --git a/jupyter/index.html b/jupyter/index.html new file mode 100644 index 0000000..792e70f --- /dev/null +++ b/jupyter/index.html @@ -0,0 +1,180 @@ + + + + + + + + + + + + + + + + + + + + + + + + + + + +
+ + + \ No newline at end of file diff --git a/jupyter/stembureau_data.ipynb b/jupyter/stembureau_data.ipynb index b12ed5f..0a82610 100644 --- a/jupyter/stembureau_data.ipynb +++ b/jupyter/stembureau_data.ipynb @@ -8,8 +8,8 @@ "\n", "Dit notebook is voor het verwerken van de data van de gemeenteraadsverkiezingen van 2022. Er zal hier stap voor stap door de data gelopen worden om het proces reproduceerbaar te maken voor latere verkiezingen. De eerste stap was de data ophalen van de bronnen, zowel de overheid als waar is mijn stemlokaal (voor geografische data van de stemlokalen). De bronnen gebruikt voor de data zijn voor de verkiezingen van 2022 is als volgt:\n", "\n", - "- [Verkiezingsuitslagen Gemeenteraad 2022](https://data.overheid.nl/dataset/verkiezingsuitslagen-gemeenteraad-2022#panel-resources)\n", - " - [Directe link naar uitslagen per gemeente CSV](https://data.overheid.nl/sites/default/files/dataset/08b04bec-3332-4c76-bb0c-68bfaeb5df43/resources/GR2022_osv4-3_2022-03-24T15.33.zip)\n", + "- [Verkiezingsuitslagen Gemeenteraad 2022](https://data.overheid.nl/dataset/08b04bec-3332-4c76-bb0c-68bfaeb5df43)\n", + " - [Directe link naar uitslagen per gemeente CSV](https://data.overheid.nl/sites/default/files/dataset/08b04bec-3332-4c76-bb0c-68bfaeb5df43/resources/GR2022_2022-03-29T15.14.zip)\n", " - [Directe link naar kandidatenlijst met uitslagen CSV](https://data.overheid.nl/sites/default/files/dataset/08b04bec-3332-4c76-bb0c-68bfaeb5df43/resources/GR2022_alle-kandidaten_2022-02-22T08.34.csv)\n", "- [Waar is mijn stemlokaal stembureau data](https://waarismijnstemlokaal.nl/data)\n", " - [Directe link naar waar is mijn stemlokaal gemeenteraad 2022 CSV (CKAN)](https://ckan.dataplatform.nl/datastore/dump/d6a1b4c4-73c8-457b-9b75-a38428bded68)\n", @@ -17,15 +17,17 @@ " - [Directe link naar GEOJSON bestand](https://data.openstate.eu/dataset/a1767f1b-bf0c-409b-b3b1-3af9954b57f4/resource/413be255-5070-48f4-b631-895097976abb/download/2022gr.geo.json)\n", "- [CBS Wijk- en buurtkaart 2021](https://www.cbs.nl/nl-nl/dossier/nederland-regionaal/geografische-data/wijk-en-buurtkaart-2021)\n", " - [Directe link naar zip bestand met SHP bestand er in](https://www.cbs.nl/-/media/cbs/dossiers/nederland-regionaal/wijk-en-buurtstatistieken/wijkbuurtkaart_2021_v1.zip)\n", - "- [CBS bevolkingsdichtheid kaart](https://www.cbs.nl/nl-nl/dossier/nederland-regionaal/geografische-data/kaart-van-100-meter-bij-100-meter-met-statistieken)\n", + "- [CBS bevolkingsdichtheid kaart 100 bij 100 meter](https://www.cbs.nl/nl-nl/dossier/nederland-regionaal/geografische-data/kaart-van-100-meter-bij-100-meter-met-statistieken)\n", " - [Directe link naar 7z bestand met SHP bestand er in](https://www.cbs.nl/-/media/cbs/dossiers/nederland-regionaal/vierkanten/100/nl_vierkant_100meter_bij_100meter.7z)\n", + "- [CBS bevolkingsdichtheid kaart 500 bij 500 meter](https://www.cbs.nl/nl-nl/dossier/nederland-regionaal/geografische-data/kaart-van-500-meter-bij-500-meter-met-statistieken)\n", + " - [Directe link naar 7z bestand met SHP bestand er in](https://www.cbs.nl/-/media/cbs/dossiers/nederland-regionaal/vierkanten/500/2021-cbs_vk500_2020_v1.zip)\n", "\n", "De eerste stap die we moeten maken is de data importeren voor de analyse, daarna kunnen we kijken hoe goed de data is, hoe we het aan kunnen vullen, en wat er mee te doen. De makkelijkste structuur die we vonden was het geojson bestand van open state en de Volkskrant, daar staan alle stembureaus al in een lijst, en we hebben een makkelijk framework om het te importeren; geopandas. We laden deze dan ook als eerste in." ] }, { "cell_type": "code", - "execution_count": 40, + "execution_count": 2, "metadata": {}, "outputs": [ { @@ -56,7 +58,7 @@ }, { "cell_type": "code", - "execution_count": 41, + "execution_count": 3, "metadata": {}, "outputs": [ { @@ -65,7 +67,7 @@ "" ] }, - "execution_count": 41, + "execution_count": 3, "metadata": {}, "output_type": "execute_result" }, @@ -89,7 +91,391 @@ }, { "cell_type": "code", - "execution_count": 36, + "execution_count": 29, + "metadata": {}, + "outputs": [ + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
GM_CODEGM_NAAMH2OOADSTEDBEV_DICHTHAANT_INWAANT_MANAANT_VROUWP_00_14_JR...P_TURKIJEP_OVER_NWOPP_TOTOPP_LANDOPP_WATERJRSTATCODEJAARShape_LengShape_Areageometry
0GM0034AlmereJA-99999999-99999999-99999999-99999999-99999999-99999999-99999999...-99999999-99999999-99999999-99999999-999999992021GM00342021122665.3586351.095623e+08MULTIPOLYGON (((150213.998 479503.726, 150087....
1GM0050ZeewoldeJA-99999999-99999999-99999999-99999999-99999999-99999999-99999999...-99999999-99999999-99999999-99999999-999999992021GM0050202183290.7736171.633050e+07MULTIPOLYGON (((170588.413 486792.192, 170570....
2GM0060AmelandJA-99999999-99999999-99999999-99999999-99999999-99999999-99999999...-99999999-99999999-99999999-99999999-999999992021GM00602021156406.8940232.085914e+08POLYGON ((196000.000 610000.000, 196000.000 60...
3GM0072HarlingenJA-99999999-99999999-99999999-99999999-99999999-99999999-99999999...-99999999-99999999-99999999-99999999-999999992021GM00722021104274.6830353.618323e+08MULTIPOLYGON (((158392.775 580357.500, 158387....
4GM0088SchiermonnikoogJA-99999999-99999999-99999999-99999999-99999999-99999999-99999999...-99999999-99999999-99999999-99999999-999999992021GM00882021152568.8589911.624101e+08POLYGON ((219000.000 616567.418, 219000.000 61...
..................................................................
430GM1966Het HogelandNEE41459947834240522378215...039030848249420592021GM19662021321301.1665764.875998e+08MULTIPOLYGON (((217037.735 601967.991, 217043....
431GM1969WesterkwartierNEE476517663678320343164417...0336887362696182021GM1969202199030.7622813.688457e+08POLYGON ((215186.661 595044.495, 215190.788 59...
432GM1970Noardeast-FryslânNEE463512045481228792260217...025164537783138622021GM19702021238426.9599263.837930e+08MULTIPOLYGON (((207769.575 603284.528, 207773....
433GM1978MolenlandenNEE393524344130223172181318...0219158181739862021GM1978202188570.3766911.915841e+08POLYGON ((123569.533 440132.167, 123576.153 44...
434GM1979EemsdeltaNEE692417045587228412274614...25364072678996182021GM19792021173966.4934522.727771e+08MULTIPOLYGON (((269190.000 594253.046, 269190....
\n", + "

435 rows × 38 columns

\n", + "
" + ], + "text/plain": [ + " GM_CODE GM_NAAM H2O OAD STED BEV_DICHTH AANT_INW \\\n", + "0 GM0034 Almere JA -99999999 -99999999 -99999999 -99999999 \n", + "1 GM0050 Zeewolde JA -99999999 -99999999 -99999999 -99999999 \n", + "2 GM0060 Ameland JA -99999999 -99999999 -99999999 -99999999 \n", + "3 GM0072 Harlingen JA -99999999 -99999999 -99999999 -99999999 \n", + "4 GM0088 Schiermonnikoog JA -99999999 -99999999 -99999999 -99999999 \n", + ".. ... ... ... ... ... ... ... \n", + "430 GM1966 Het Hogeland NEE 414 5 99 47834 \n", + "431 GM1969 Westerkwartier NEE 476 5 176 63678 \n", + "432 GM1970 Noardeast-Fryslân NEE 463 5 120 45481 \n", + "433 GM1978 Molenlanden NEE 393 5 243 44130 \n", + "434 GM1979 Eemsdelta NEE 692 4 170 45587 \n", + "\n", + " AANT_MAN AANT_VROUW P_00_14_JR ... P_TURKIJE P_OVER_NW OPP_TOT \\\n", + "0 -99999999 -99999999 -99999999 ... -99999999 -99999999 -99999999 \n", + "1 -99999999 -99999999 -99999999 ... -99999999 -99999999 -99999999 \n", + "2 -99999999 -99999999 -99999999 ... -99999999 -99999999 -99999999 \n", + "3 -99999999 -99999999 -99999999 ... -99999999 -99999999 -99999999 \n", + "4 -99999999 -99999999 -99999999 ... -99999999 -99999999 -99999999 \n", + ".. ... ... ... ... ... ... ... \n", + "430 24052 23782 15 ... 0 3 90308 \n", + "431 32034 31644 17 ... 0 3 36887 \n", + "432 22879 22602 17 ... 0 2 51645 \n", + "433 22317 21813 18 ... 0 2 19158 \n", + "434 22841 22746 14 ... 2 5 36407 \n", + "\n", + " OPP_LAND OPP_WATER JRSTATCODE JAAR Shape_Leng Shape_Area \\\n", + "0 -99999999 -99999999 2021GM0034 2021 122665.358635 1.095623e+08 \n", + "1 -99999999 -99999999 2021GM0050 2021 83290.773617 1.633050e+07 \n", + "2 -99999999 -99999999 2021GM0060 2021 156406.894023 2.085914e+08 \n", + "3 -99999999 -99999999 2021GM0072 2021 104274.683035 3.618323e+08 \n", + "4 -99999999 -99999999 2021GM0088 2021 152568.858991 1.624101e+08 \n", + ".. ... ... ... ... ... ... \n", + "430 48249 42059 2021GM1966 2021 321301.166576 4.875998e+08 \n", + "431 36269 618 2021GM1969 2021 99030.762281 3.688457e+08 \n", + "432 37783 13862 2021GM1970 2021 238426.959926 3.837930e+08 \n", + "433 18173 986 2021GM1978 2021 88570.376691 1.915841e+08 \n", + "434 26789 9618 2021GM1979 2021 173966.493452 2.727771e+08 \n", + "\n", + " geometry \n", + "0 MULTIPOLYGON (((150213.998 479503.726, 150087.... \n", + "1 MULTIPOLYGON (((170588.413 486792.192, 170570.... \n", + "2 POLYGON ((196000.000 610000.000, 196000.000 60... \n", + "3 MULTIPOLYGON (((158392.775 580357.500, 158387.... \n", + "4 POLYGON ((219000.000 616567.418, 219000.000 61... \n", + ".. ... \n", + "430 MULTIPOLYGON (((217037.735 601967.991, 217043.... \n", + "431 POLYGON ((215186.661 595044.495, 215190.788 59... \n", + "432 MULTIPOLYGON (((207769.575 603284.528, 207773.... \n", + "433 POLYGON ((123569.533 440132.167, 123576.153 44... \n", + "434 MULTIPOLYGON (((269190.000 594253.046, 269190.... \n", + "\n", + "[435 rows x 38 columns]" + ] + }, + "execution_count": 29, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "nl_map" + ] + }, + { + "cell_type": "code", + "execution_count": 4, "metadata": {}, "outputs": [ { @@ -98,7 +484,7 @@ "" ] }, - "execution_count": 36, + "execution_count": 4, "metadata": {}, "output_type": "execute_result" }, @@ -132,7 +518,7 @@ }, { "cell_type": "code", - "execution_count": 25, + "execution_count": 5, "metadata": {}, "outputs": [ { @@ -925,14 +1311,14 @@ }, { "cell_type": "code", - "execution_count": 30, + "execution_count": 6, "metadata": {}, "outputs": [ { "name": "stderr", "output_type": "stream", "text": [ - "/tmp/ipykernel_39610/291552514.py:3: UserWarning: Geometry is in a geographic CRS. Results from 'centroid' are likely incorrect. Use 'GeoSeries.to_crs()' to re-project geometries to a projected CRS before this operation.\n", + "/tmp/ipykernel_15650/291552514.py:3: UserWarning: Geometry is in a geographic CRS. Results from 'centroid' are likely incorrect. Use 'GeoSeries.to_crs()' to re-project geometries to a projected CRS before this operation.\n", "\n", " mobiel_mask = df_geojson['geometry'].centroid.x == 0\n" ] @@ -1021,131 +1407,13 @@ }, { "cell_type": "code", - "execution_count": 34, + "execution_count": 7, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - " Stembureau Adres Locatie \\\n", - "0 1 9461BH SB1 \n", - "1 2 9461DA SB2 \n", - "2 3 9461JA SB3 \n", - "3 4 9451KD SB4 \n", - "4 6 9454PL SB6 \n", - "... ... ... ... \n", - "12591 703 5391AR SB703 \n", - "12592 705 5391AR SB705 \n", - "12593 750 5382KE SB750 \n", - "12594 751 5382KE SB751 \n", - "12595 752 5283KE SB752 \n", - "\n", - " description Geldige stemmen \\\n", - "0 Stembureau Gemeentehuis Gieten (postcode: 9461... 784 \n", - "1 Stembureau OBS Gieten (postcode: 9461 DA) 562 \n", - "2 Stembureau Zorgcentrum Dekelhem (postcode: 946... 566 \n", - "3 Stembureau Ontmoetingscentrum Boerhorn Rolde (... 1495 \n", - "4 Stembureau Dropshuis de Eekhof (postcode: 9454... 347 \n", - "... ... ... \n", - "12591 Stembureau Stembureau Gemeenschapshuis de Meen... 268 \n", - "12592 Stembureau Stembureau Gemeenschapshuis De Meen... 398 \n", - "12593 Stembureau Stembureau Gemeenschapshuis 't Zijl... 663 \n", - "12594 Stembureau Stembureau Gemeenschapshuis 't Zijl... 170 \n", - "12595 Stembureau Stembureau Gemeenschapshuis 't Zijl... 222 \n", - "\n", - " Opgeroepen Ongeldig Blanco Geldige stempassen \\\n", - "0 2780 3 3 700 \n", - "1 1396 0 0 518 \n", - "2 1409 2 2 516 \n", - "3 2209 2 4 1335 \n", - "4 477 0 2 298 \n", - "... ... ... ... ... \n", - "12591 0 0 1 237 \n", - "12592 0 1 0 359 \n", - "12593 2321 2 0 552 \n", - "12594 0 0 0 151 \n", - "12595 0 0 0 193 \n", - "\n", - " Geldige volmachtbewijzen ... \\\n", - "0 90 ... \n", - "1 44 ... \n", - "2 54 ... \n", - "3 166 ... \n", - "4 51 ... \n", - "... ... ... \n", - "12591 32 ... \n", - "12592 40 ... \n", - "12593 113 ... \n", - "12594 19 ... \n", - "12595 29 ... \n", - "\n", - " Nationale Bond tegen Overheidszaken - DH Haags Belang INL Den Haag \\\n", - "0 NaN NaN NaN \n", - "1 NaN NaN NaN \n", - "2 NaN NaN NaN \n", - "3 NaN NaN NaN \n", - "4 NaN NaN NaN \n", - "... ... ... ... \n", - "12591 NaN NaN NaN \n", - "12592 NaN NaN NaN \n", - "12593 NaN NaN NaN \n", - "12594 NaN NaN NaN \n", - "12595 NaN NaN NaN \n", - "\n", - " Rosmalens Belang De Bossche Groenen \\\n", - "0 NaN NaN \n", - "1 NaN NaN \n", - "2 NaN NaN \n", - "3 NaN NaN \n", - "4 NaN NaN \n", - "... ... ... \n", - "12591 105.0 2.0 \n", - "12592 174.0 6.0 \n", - "12593 47.0 19.0 \n", - "12594 8.0 11.0 \n", - "12595 10.0 2.0 \n", - "\n", - " \"Leefbaar 's-Hertogenbosch\" Paul Kagie \\\n", - "0 NaN \n", - "1 NaN \n", - "2 NaN \n", - "3 NaN \n", - "4 NaN \n", - "... ... \n", - "12591 6.0 \n", - "12592 2.0 \n", - "12593 53.0 \n", - "12594 13.0 \n", - "12595 22.0 \n", - "\n", - " RAADSGROEPERING ''BOSCH-BELANG'' gewoon ge-DREVEN \\\n", - "0 NaN NaN \n", - "1 NaN NaN \n", - "2 NaN NaN \n", - "3 NaN NaN \n", - "4 NaN NaN \n", - "... ... ... \n", - "12591 0.0 33.0 \n", - "12592 3.0 31.0 \n", - "12593 2.0 62.0 \n", - "12594 1.0 22.0 \n", - "12595 2.0 28.0 \n", - "\n", - " VOOR Den Bosch Joep Gersjes geometry \n", - "0 NaN POINT (6.75899 53.00524) \n", - "1 NaN POINT (6.75990 52.99975) \n", - "2 NaN POINT (6.76600 53.00494) \n", - "3 NaN POINT (6.64736 52.98281) \n", - "4 NaN POINT (6.60459 52.95269) \n", - "... ... ... \n", - "12591 0.0 POINT (5.43290 51.72810) \n", - "12592 2.0 POINT (5.43290 51.72810) \n", - "12593 5.0 POINT (5.45919 51.70595) \n", - "12594 1.0 POINT (5.45919 51.70595) \n", - "12595 0.0 POINT (5.45919 51.70595) \n", - "\n", - "[12532 rows x 973 columns]\n", " Stembureau Adres Locatie \\\n", "2520 32 SB32 \n", "2521 33 SB33 \n", @@ -1275,7 +1543,7 @@ }, { "cell_type": "code", - "execution_count": 37, + "execution_count": 8, "metadata": {}, "outputs": [ { @@ -1284,7 +1552,7 @@ "" ] }, - "execution_count": 37, + "execution_count": 8, "metadata": {}, "output_type": "execute_result" }, @@ -1316,7 +1584,7 @@ }, { "cell_type": "code", - "execution_count": 39, + "execution_count": 9, "metadata": {}, "outputs": [ { @@ -1325,13 +1593,13 @@ "" ] }, - "execution_count": 39, + "execution_count": 9, "metadata": {}, "output_type": "execute_result" }, { "data": { - "image/png": "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", + "image/png": "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", "text/plain": [ "
" ] @@ -1343,45 +1611,270 @@ } ], "source": [ - "nl_map_dst = gpd.read_file(r'../data/shape/Netherlands_shapefile/NL_vierkant100m.shp')\n", - "nl_map_dst.to_crs(epsg=4326).plot(color = green)" + "nl_map_cbs = gpd.read_file(r'../data/shape/Netherlands_shapefile/CBS_vk500_2020_v1.shp')\n", + "nl_map_cbs.to_crs(epsg=4326).plot()" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "Het verwerken van de 500m bij 500m vierkanten kaar duurde heel lang (langer dan een half uur) op de laptop, en is dus niet ideaal tenzij we dit niveau van detail écht nodig hebben. Er is namelijk nog een andere kaart met dezelfde gegevens beschikbaar, maar met een grid van 500 bij 500 meter in plaats van 100 bij 100. Laten we dus kijken hoe vaak een stemlokaal dichterbij dan 100 meter van de dichtstbijzijnde andere is. De makkelijkste manier om dat te doen zonder alle punten met alle andere te vergelijken (heel veel moeite), is een extra dataframe maken als een kopie, alle indexen 1 opschuiven (want alle stemlokalen zijn al in een volgorde van clustering), en dan de laagste afstand bekijken." ] }, { "cell_type": "code", - "execution_count": 47, + "execution_count": 10, "metadata": {}, "outputs": [ { "name": "stdout", "output_type": "stream", "text": [ - " C28992R100 geometry\n", - "0 E2266N6194 POLYGON ((226600.000 619400.000, 226600.000 61...\n", - "1 E2267N6194 POLYGON ((226700.000 619500.000, 226800.000 61...\n", - "2 E2268N6194 POLYGON ((226800.000 619500.000, 226900.000 61...\n", - "3 E2269N6194 POLYGON ((226900.000 619500.000, 227000.000 61...\n", - "4 E2270N6194 POLYGON ((227000.000 619500.000, 227100.000 61...\n", - "... ... ...\n", - "3723283 E1927N3068 POLYGON ((192700.000 306900.000, 192800.000 30...\n", - "3723284 E1921N3067 POLYGON ((192100.000 306800.000, 192200.000 30...\n", - "3723285 E1922N3067 POLYGON ((192200.000 306800.000, 192300.000 30...\n", - "3723286 E1923N3067 POLYGON ((192300.000 306800.000, 192400.000 30...\n", - "3723287 E1924N3067 POLYGON ((192400.000 306800.000, 192500.000 30...\n", - "\n", - "[3723288 rows x 2 columns]\n" + "11165 0.002608\n", + "1021 0.004853\n", + "3196 0.004875\n", + "1923 2.693202\n", + "3757 3.609362\n", + " ... \n", + "7444 253385.654588\n", + "9711 263408.187856\n", + "4392 266925.200994\n", + "9297 271954.076837\n", + "0 NaN\n", + "Name: distance, Length: 11670, dtype: float64\n" ] } ], "source": [ - "print(nl_map_dst)" + "df_shifted = df_geojson_clean.to_crs('EPSG:28992')\n", + "\n", + "df_shifted['geometry (shifted)'] = df_shifted['geometry'].shift(periods=1)\n", + "df_shifted['distance'] = df_shifted['geometry'].distance(df_shifted['geometry (shifted)'])\n", + "df_shifted.sort_values(['distance'], inplace=True, ascending=True)\n", + "zero_mask_booth = df_shifted['distance'] == 0.000000\n", + "df_shifted = df_shifted[~zero_mask_booth]\n", + "print(df_shifted['distance'])" + ] + }, + { + "cell_type": "markdown", + "metadata": {}, + "source": [ + "We kunnen dus vaststellen dat afstanden van onder de 500 meter waarschijnlijk zeldzaam zijn, en we verder kunnen gaan met de 500 meter bij 500 meter kaart. Laten we nu dus een projectie proberen te maken met de bevolkingsdichtheid erop om het te vergelijken met de stemlokalen en hun posities en clustering." ] }, { "cell_type": "code", - "execution_count": null, + "execution_count": 11, + "metadata": {}, + "outputs": [ + { + "name": "stdout", + "output_type": "stream", + "text": [ + "['c28992r500', 'INWONER', 'MAN', 'VROUW', 'INW_014', 'INW_1524', 'INW_2544', 'INW_4564', 'INW_65PL', 'P_NL_ACHTG', 'P_WE_MIG_A', 'P_NW_MIG_A', 'AANTAL_HH', 'TOTHH_EENP', 'TOTHH_MPZK', 'HH_EENOUD', 'HH_TWEEOUD', 'GEM_HH_GR', 'WONING', 'WONVOOR45', 'WON_4564', 'WON_6574', 'WON_7584', 'WON_8594', 'WON_9504', 'WON_0514', 'WON_1524', 'WON_MRGEZ', 'P_KOOPWON', 'P_HUURWON', 'WON_HCORP', 'WON_NBEW', 'WOZWONING', 'UITKMINAOW', 'OAD', 'STED', 'geometry']\n" + ] + }, + { + "data": { + "text/html": [ + "
\n", + "\n", + "\n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + " \n", + "
c28992r500INWONERgeometry
267E2050N61105POLYGON ((205000.000 611500.000, 205500.000 61...
269E2060N611015POLYGON ((206000.000 611500.000, 206500.000 61...
292E2055N610520POLYGON ((205500.000 611000.000, 206000.000 61...
293E2060N6105185POLYGON ((206000.000 611000.000, 206500.000 61...
294E2065N6105340POLYGON ((206500.000 611000.000, 207000.000 61...
............
151071E1970N307510POLYGON ((197000.000 308000.000, 197500.000 30...
151073E1980N307590POLYGON ((198000.000 308000.000, 198500.000 30...
151074E1985N307515POLYGON ((198500.000 308000.000, 199000.000 30...
151100E1980N30705POLYGON ((198000.000 307500.000, 198500.000 30...
151106E1920N306520POLYGON ((192000.000 307000.000, 192500.000 30...
\n", + "

70656 rows × 3 columns

\n", + "
" + ], + "text/plain": [ + " c28992r500 INWONER geometry\n", + "267 E2050N6110 5 POLYGON ((205000.000 611500.000, 205500.000 61...\n", + "269 E2060N6110 15 POLYGON ((206000.000 611500.000, 206500.000 61...\n", + "292 E2055N6105 20 POLYGON ((205500.000 611000.000, 206000.000 61...\n", + "293 E2060N6105 185 POLYGON ((206000.000 611000.000, 206500.000 61...\n", + "294 E2065N6105 340 POLYGON ((206500.000 611000.000, 207000.000 61...\n", + "... ... ... ...\n", + "151071 E1970N3075 10 POLYGON ((197000.000 308000.000, 197500.000 30...\n", + "151073 E1980N3075 90 POLYGON ((198000.000 308000.000, 198500.000 30...\n", + "151074 E1985N3075 15 POLYGON ((198500.000 308000.000, 199000.000 30...\n", + "151100 E1980N3070 5 POLYGON ((198000.000 307500.000, 198500.000 30...\n", + "151106 E1920N3065 20 POLYGON ((192000.000 307000.000, 192500.000 30...\n", + "\n", + "[70656 rows x 3 columns]" + ] + }, + "execution_count": 11, + "metadata": {}, + "output_type": "execute_result" + } + ], + "source": [ + "print(list(nl_map_cbs))\n", + "nl_map_dst = nl_map_cbs.drop(columns=['MAN', 'VROUW', 'INW_014', 'INW_1524', 'INW_2544', 'INW_4564', 'INW_65PL', 'P_NL_ACHTG', 'P_WE_MIG_A', 'P_NW_MIG_A', 'AANTAL_HH', 'TOTHH_EENP', 'TOTHH_MPZK', 'HH_EENOUD', 'HH_TWEEOUD', 'GEM_HH_GR', 'WONING', 'WONVOOR45', 'WON_4564', 'WON_6574', 'WON_7584', 'WON_8594', 'WON_9504', 'WON_0514', 'WON_1524', 'WON_MRGEZ', 'P_KOOPWON', 'P_HUURWON', 'WON_HCORP', 'WON_NBEW', 'WOZWONING', 'UITKMINAOW', 'OAD', 'STED'])\n", + "zero_mask_pop = nl_map_dst['INWONER'] == -99997\n", + "nl_map_dst = nl_map_dst[~zero_mask_pop]\n", + "nl_map_dst" + ] + }, + { + "cell_type": "code", + "execution_count": 46, "metadata": {}, "outputs": [], + "source": [ + "import folium\n", + "from folium import plugins\n", + "\n", + "map = folium.Map(location = [52.155, 5.3875], zoom_start = 9, tiles=\"cartodbdark_matter\", prefer_canvas=True)\n", + "\n", + "nl_map_dst.to_crs(epsg=4326)\n", + "#Make sure the index is a string so folium can read it correctly as a key.\n", + "nl_map_dst['c28992r500'] = nl_map_dst['c28992r500'].apply(lambda x: str(x))\n", + "\n", + "folium.Choropleth(\n", + " geo_data = nl_map_dst,\n", + " name=\"Bevolkingsdichtheid\",\n", + " data = nl_map_dst,\n", + " columns = [\"c28992r500\", \"INWONER\"],\n", + " key_on = 'feature.properties.c28992r500',\n", + " fill_color = 'RdPu',\n", + " nan_fill_color= 'white',\n", + " fill_opacity = 0.7,\n", + " nan_fill_opacity = 0.7,\n", + " line_opacity = 0,\n", + " legend_name = 'Bevolkingsdichtheid',\n", + " smooth_factor = 1.0,\n", + " show=False\n", + ").add_to(map)\n", + "\n", + "# Renders the map to an HTML file and displays it in an embed.\n", + "def embed_map(m):\n", + " #from IPython.display import IFrame\n", + " m.save('index.html')\n", + " #return IFrame('index.html', width='100%', height='750px')" + ] + }, + { + "cell_type": "code", + "execution_count": 45, + "metadata": {}, + "outputs": [], + "source": [ + "df_geojson_clean_hmp = [[point.xy[1][0], point.xy[0][0]] for point in df_geojson_clean.geometry]\n", + "\n", + "plugins.HeatMap(df_geojson_clean_hmp, name=\"Stemlokalen Heatmap\").add_to(map)\n", + "\n", + "folium.LayerControl().add_to(map)\n", + "\n", + "\n", + "\n", + "embed_map(map)\n", + "\n", + "\n" + ] + }, + { + "cell_type": "code", + "execution_count": 13, + "metadata": {}, + "outputs": [ + { + "ename": "SyntaxError", + "evalue": "invalid syntax (745234620.py, line 3)", + "output_type": "error", + "traceback": [ + "\u001b[0;36m Input \u001b[0;32mIn [13]\u001b[0;36m\u001b[0m\n\u001b[0;31m dfwims.\u001b[0m\n\u001b[0m ^\u001b[0m\n\u001b[0;31mSyntaxError\u001b[0m\u001b[0;31m:\u001b[0m invalid syntax\n" + ] + } + ], "source": [ "df_wims = pd.read_csv(r'../data/wims.csv')\n", "print(df_wims)\n", diff --git a/jupyter/table.png b/jupyter/table.png new file mode 100644 index 0000000..ddffcb6 Binary files /dev/null and b/jupyter/table.png differ