def test_points():
x = np.arange(10).astype(np.float)
y = np.arange(10).astype(np.float)**2
points = points_from_xy(x, y)
assert isinstance(points, GeometryArray)
for i in range(10):
assert isinstance(points[i], shapely.geometry.Point)
assert points[i].x == x[i]
assert points[i].y == y[i]
def load_metro_areas_shapefile():
shpfile_path = (
CWD /
"USA_Core_Based_Statistical_Area" # / "USA_Core_Based_Statistical_Area.shp"
)
metro_areas = gpd.read_file(shpfile_path)
metro_areas = metro_areas.to_crs(epsg=4326)
corrected_metro_centroids = pd.read_csv(CWD.parent / "bin" /
"msa_urban_centroids.csv")
corrected_metro_centroids["CBSA_ID"] = corrected_metro_centroids[
"CBSA_ID"].astype("str")
corrected_metro_centroids = corrected_metro_centroids.set_index("CBSA_ID")
corrected_metro_centroids = gpd.GeoDataFrame(
corrected_metro_centroids,
geometry=points_from_xy(
corrected_metro_centroids["msa_longitude"],
corrected_metro_centroids["msa_latitude"],
),
crs="EPSG:4326",
)
metro_areas["center"] = find_centroid(metro_areas)
metro_areas["corrected_center"] = metro_areas["CBSA_ID"].map(
corrected_metro_centroids["geometry"])
metro_areas["msa_center"] = metro_areas["center"]
metro_areas.loc[~metro_areas["corrected_center"].isna(),
"center"] = metro_areas["corrected_center"]
keep_cols = [
"CBSA_ID",
"NAME",
"CBSA_TYPE",
"POPULATION",
"center",
"msa_center",
"geometry",
]
# metro_areas["geometry"] = metro_areas["center"]
metro_areas = metro_areas.loc[:, keep_cols]
metro_areas["metro_id"] = metro_areas["CBSA_ID"]
metro_areas.columns = metro_areas.columns.str.lower()
metro_areas["state"] = metro_areas["name"].str.split(", ").str[-1]
metro_areas = metro_areas.loc[
~metro_areas.state.isin(["AK", "HI", "PR"]), :]
NY_Z_J_lon_lat = (-73.930488, 40.695448)
NY_Z_K_lon_lat = (-73.008906, 40.840391)
extra_metros = pd.DataFrame([["NY_Z_J", 1e6], ["NY_Z_K", 1e6]],
columns=["metro_id", "population"])
extra_metros = gpd.GeoDataFrame(
extra_metros,
geometry=points_from_xy(*zip(NY_Z_J_lon_lat, NY_Z_K_lon_lat)),
crs="EPSG:4326",
)
extra_metros["center"] = extra_metros["geometry"]
metro_areas = pd.concat([metro_areas, extra_metros],
ignore_index=True,
sort=False)
return metro_areas
