Source: https://www.kaggle.com/learn/geospatial-analysis,
This is the code summary note for the Kaggle geospatial analysis
geopandas frame has a geometry column that shows either a point/line/polygon
Part 1: Your first map
source: https://www.kaggle.com/alexisbcook/your-first-map
import geopandas as gpd
from learntools.core import binder
binder.bind(globals())
from learntools.geospatial.ex1 import *
### LOAD DATA ###
# shapefile
shapef = gpd.read_file(loans_sahpe_filepath)
# shpae file comes with crs (coordinate reference)
print(regions.crs)
# kml data
gpd.io.file.fiona.drvsupport.supported_drivers['KML'] = 'rw'
kmlf = gpd.read_file("mykmldata.kml", driver='KML')
# geopandas dataset
world_filepath = gpd.datasets.get_path('naturalearth_lowres')
gpdf = gpd.read_file(world_filepath)
### MAPPING
# plotting the basemap
ax = world.plot(figsize=(20,20), color='whitesmoke', linestyle=':', edgecolor='black')
# adding your data on the basedmap ax
shapef.plot(color='black', markersize=2, ax=ax)
# subsetting data
subshapef = shapef.loc[world_loans.country == 'Philippines'].copy()Part II – Coordinate Reference Systems
Source: https://www.kaggle.com/alexisbcook/coordinate-reference-systems
# shapefile comes with the crs
shapef = gpd.read_file(loans_sahpe_filepath)
print(shapef.crs)
# need to set crs for csv file load that has lon and lat
csvf = pd.read_csv("mycsvfile.csv")
# converts the csv dataframe to geodataframe
gdf = gpd.GeoDataFrame(csvf, geometry=gdp.points_from_xy(csvf.lon, csvf.lat))
# manually set the crs to EPSG 4326
gdf.crs = {'init': 'epsg:4326'}
# reprojecting with differetn crs
ax = world.plot(figsize=(8,8), color='whitesmoke', linestyle=':', edgecolor='black')
gdf.to_crs(epsg=32630).plot(markersize=1, ax=ax)
# in case EPSG code is not available in geopandas, here's a hack example
# Change the CRS to EPSG 4326
gdf.to_crs("+proj=longlat +ellps=WGS84 +datum=WGS84 +no_defs").head()
# accessing geometry attributes
# Get the x-coordinate of each point
gdf.geometry.head().x
gdf.geometry.head().y
# Get the area attribute from geometry
gdf.geometry.area
# Add a column "AREA" in the gdf dataframe
gdf.loc[:, "AREA"] = gdf.geometry.area / 10**6 #converting to square meters
print("Area of entire country in gdf: {} square kilometers".format(gdf.AREA.sum())) #adding all rows of AREA
print("CRS:", gdf.crs)
gdf.head()
