def network(fd: gpd.GeoDataFrame,
modes=["bike"],
max_rows=1000) -> gpd.GeoDataFrame:
r"""
Reduce route data to a network
This function reduces route data in GeoDataFrame `fd` to a network for the
modes of transporation listed in `modes`. All line segments of routes that
overlap are reduced to one segment and their mode numbers are summed up.
Parameters
----------
modes : list of str, defaults to ["bike"]
List of modes of transportation that the network is computed for.
Defaults to ["bike"].
max_rows : int, defaults to 4000
To reduce the memory footprint a GeoDataFrames is split up in blocks of
`max_rows` rows. This value can be increased on computers with enough
RAM.
Returns
-------
geopandas.GeoDataFrame
GeoDataFrame containing the network.
See Also
--------
~stplanpy.cycle.routes
Examples
--------
.. code-block:: python
mport pandas as pd
port geopandas as gpd
om shapely import wkt
om stplanpy import route
Create DataFrames
df = pd.DataFrame(
{"all": [4, 3, 2, 5],
"bike": [2, 0, 1, 3],
"go_dutch": [3, 5, 0, 4],
"geometry": ["LINESTRING(1 0,0 0,1 1,2 1,3 0)",
"LINESTRING(0 2,1 1,2 1,3 2,2 2)",
"LINESTRING(1 0,1 1,2 1,2 0)",
"LINESTRING(1 2,1 1,2 1,2 2,3 2)"]})
# Convert to WTK
df["geometry"] = gpd.GeoSeries.from_wkt(df["geometry"])
# Create GeoDataFrame
gdf = gpd.GeoDataFrame(df, geometry='geometry')
# Compute the network
network = gdf.network(modes=["bike", "go_dutch"])
"""
if (fd.shape[0] < max_rows):
return fd._networks(modes=modes)
else:
# Compute maximum number of GeoDataFrames
max_frames = int(fd.shape[0] / max_rows) + 1
# Create empty list
frames = []
# Iterate over all GeoDataFrames
for idx in range(max_frames):
idx0 = idx * max_rows
idx1 = (idx + 1) * max_rows - 1
if (idx1 > fd.shape[0]):
idx1 = fd.shape[0]
frames.append(fd.iloc[idx0:idx1]._networks(modes=modes))
# Combine all GeoDataFrames
return pd.concat(frames, ignore_index=True).explode(
ignore_index=True)._networks(modes=modes)
