Skip to content

Commit

Permalink
ST data preprocessing indent issue resolved
Browse files Browse the repository at this point in the history
  • Loading branch information
@kanchanchy
kanchanchy committed Jul 31, 2023
1 parent 6b4c666 commit 9b36e07
Showing 1 changed file with 135 additions and 135 deletions.
270 changes: 135 additions & 135 deletions geotorchai/preprocessing/grid/st_manager.py
View file
Original file line number Diff line number Diff line change
Expand Up @@ -18,141 +18,141 @@

class STManager:

@classmethod
def get_grid_cell_polygons(cls, geo_df: DataFrame, geometry: str, partitions_x: int, partitions_y: int):
'''
Function generates a grid of partitions_x times partitions_y cells
partitions_x cells along the latitude and partitions_y cells along the longitude
Parameters
...........
geo_df: pyspark dataframe containing a column of geometry type
geometry: name of the geometry typed column in geo_df dataframe
partitions_x: number of partitions along latitude
partitions_y: number of partitions along longitude
Returns
........
a pyspark dataframe constsiting of two columns: id of each cell and polygon object representing each cell
'''

# retrieve the SparkSession instance
spark = SparkRegistration._get_spark_session()

geo_df.createOrReplaceTempView("geo_df")
boundary = \
spark.sql("SELECT ST_Envelope_Aggr(geo_df.{0}) as boundary FROM geo_df".format(geometry)).collect()[0][0]
x_arr, y_arr = boundary.exterior.coords.xy
x = list(x_arr)
y = list(y_arr)

min_x, min_y, max_x, max_y = min(x), min(y), max(x), max(y)
interval_x = (max_x - min_x) / partitions_x
interval_y = (max_y - min_y) / partitions_y

polygons = []
for i in range(partitions_y):
for j in range(partitions_x):
polygons.append(Polygon([[min_x + interval_x * j, min_y + interval_y * i],
[min_x + interval_x * (j + 1), min_y + interval_y * i],
[min_x + interval_x * (j + 1), min_y + interval_y * (i + 1)],
[min_x + interval_x * j, min_y + interval_y * (i + 1)],
[min_x + interval_x * j, min_y + interval_y * i]]))
return polygons


@classmethod
def getHexagonalLayer(cls, df, col_lat, col_lon, fraction):
pd_df = df.sample(fraction).select(*[col_lon, col_lat]).toPandas()
lat_mean = (40.491370 + 40.91553) / 2
lon_mean = (-74.259090 + -73.700180) / 2

layer = pdk.Layer(
'HexagonLayer', # `type` positional argument is here
pd_df,
get_position=[col_lon, col_lat],
auto_highlight=True,
elevation_scale=50,
pickable=True,
elevation_range=[0, 3000],
extruded=True,
coverage=1)

# Set the viewport location
view_state = pdk.ViewState(
longitude=lon_mean,
latitude=lat_mean,
zoom=8,
min_zoom=5,
max_zoom=15,
pitch=40.5,
bearing=-27.36)

# Combined all of it and render a viewport
r = pdk.Deck(layers=[layer], initial_view_state=view_state)
return r


@classmethod
def getStGridLayer(cls, df, timestamp_index, col_timestamp, col_geometry, col_feature, col_id, polygons):
df_geo = df.filter(df[col_timestamp] == timestamp_index)

feature_objects = []
for i in range(len(polygons)):
feature_objects.append(gjson.Feature(geometry=polygons[i], properties={"pickup_count": 0}))

max_value = 1
for row in df_geo.collect():
# geom = geojson.Feature(geometry=row['geometry'], properties={"pickup_count": row['aggregated_feature']}).geometry
feature_objects[row[col_id]] = gjson.Feature(geometry=polygons[row[col_id]], properties={"pickup_count": row[col_feature]})
if row[col_feature] > max_value:
max_value = row[col_feature]

feature_collection = gjson.FeatureCollection(feature_objects)
# Convert the FeatureCollection to a string in GeoJSON format
geojson_data = gjson.dumps(feature_collection, sort_keys=True)
geojson_data = json.loads(geojson_data)

LAND_COVER = [[-74.259090, 40.491370], [-74.259090, 40.91553], [-73.700180, 40.91553], [-73.700180, 40.491370]]

lat_mean = (40.491370 + 40.91553) / 2
lon_mean = (-74.259090 + -73.700180) / 2

INITIAL_VIEW_STATE = pdk.ViewState(
longitude=lon_mean,
latitude=lat_mean,
zoom=9,
max_zoom=16,
pitch=45,
bearing=0
)

polygon = pdk.Layer(
'PolygonLayer',
LAND_COVER,
stroked=False,
# processes the data as a flat longitude-latitude pair
get_polygon='-',
get_fill_color=[0, 0, 0, 20]
)

geojson = pdk.Layer(
'GeoJsonLayer',
geojson_data,
opacity=0.2,
stroked=False,
filled=True,
extruded=True,
wireframe=True,
get_elevation='properties.pickup_count*75',
get_fill_color='[255, 255, properties.pickup_count*255/max_value]',
get_line_color=[255, 255, 255]
)

r = pdk.Deck(
layers=[polygon, geojson],
initial_view_state=INITIAL_VIEW_STATE)
return r
@classmethod
def get_grid_cell_polygons(cls, geo_df: DataFrame, geometry: str, partitions_x: int, partitions_y: int):
'''
Function generates a grid of partitions_x times partitions_y cells
partitions_x cells along the latitude and partitions_y cells along the longitude
Parameters
...........
geo_df: pyspark dataframe containing a column of geometry type
geometry: name of the geometry typed column in geo_df dataframe
partitions_x: number of partitions along latitude
partitions_y: number of partitions along longitude
Returns
........
a pyspark dataframe constsiting of two columns: id of each cell and polygon object representing each cell
'''

# retrieve the SparkSession instance
spark = SparkRegistration._get_spark_session()

geo_df.createOrReplaceTempView("geo_df")
boundary = \
spark.sql("SELECT ST_Envelope_Aggr(geo_df.{0}) as boundary FROM geo_df".format(geometry)).collect()[0][0]
x_arr, y_arr = boundary.exterior.coords.xy
x = list(x_arr)
y = list(y_arr)

min_x, min_y, max_x, max_y = min(x), min(y), max(x), max(y)
interval_x = (max_x - min_x) / partitions_x
interval_y = (max_y - min_y) / partitions_y

polygons = []
for i in range(partitions_y):
for j in range(partitions_x):
polygons.append(Polygon([[min_x + interval_x * j, min_y + interval_y * i],
[min_x + interval_x * (j + 1), min_y + interval_y * i],
[min_x + interval_x * (j + 1), min_y + interval_y * (i + 1)],
[min_x + interval_x * j, min_y + interval_y * (i + 1)],
[min_x + interval_x * j, min_y + interval_y * i]]))
return polygons


@classmethod
def getHexagonalLayer(cls, df, col_lat, col_lon, fraction):
pd_df = df.sample(fraction).select(*[col_lon, col_lat]).toPandas()
lat_mean = (40.491370 + 40.91553) / 2
lon_mean = (-74.259090 + -73.700180) / 2

layer = pdk.Layer(
'HexagonLayer', # `type` positional argument is here
pd_df,
get_position=[col_lon, col_lat],
auto_highlight=True,
elevation_scale=50,
pickable=True,
elevation_range=[0, 3000],
extruded=True,
coverage=1)

# Set the viewport location
view_state = pdk.ViewState(
longitude=lon_mean,
latitude=lat_mean,
zoom=8,
min_zoom=5,
max_zoom=15,
pitch=40.5,
bearing=-27.36)

# Combined all of it and render a viewport
r = pdk.Deck(layers=[layer], initial_view_state=view_state)
return r


@classmethod
def getStGridLayer(cls, df, timestamp_index, col_timestamp, col_geometry, col_feature, col_id, polygons):
df_geo = df.filter(df[col_timestamp] == timestamp_index)

feature_objects = []
for i in range(len(polygons)):
feature_objects.append(gjson.Feature(geometry=polygons[i], properties={"pickup_count": 0}))

max_value = 1
for row in df_geo.collect():
# geom = geojson.Feature(geometry=row['geometry'], properties={"pickup_count": row['aggregated_feature']}).geometry
feature_objects[row[col_id]] = gjson.Feature(geometry=polygons[row[col_id]], properties={"pickup_count": row[col_feature]})
if row[col_feature] > max_value:
max_value = row[col_feature]

feature_collection = gjson.FeatureCollection(feature_objects)
# Convert the FeatureCollection to a string in GeoJSON format
geojson_data = gjson.dumps(feature_collection, sort_keys=True)
geojson_data = json.loads(geojson_data)

LAND_COVER = [[-74.259090, 40.491370], [-74.259090, 40.91553], [-73.700180, 40.91553], [-73.700180, 40.491370]]

lat_mean = (40.491370 + 40.91553) / 2
lon_mean = (-74.259090 + -73.700180) / 2

INITIAL_VIEW_STATE = pdk.ViewState(
longitude=lon_mean,
latitude=lat_mean,
zoom=9,
max_zoom=16,
pitch=45,
bearing=0
)

polygon = pdk.Layer(
'PolygonLayer',
LAND_COVER,
stroked=False,
# processes the data as a flat longitude-latitude pair
get_polygon='-',
get_fill_color=[0, 0, 0, 20]
)

geojson = pdk.Layer(
'GeoJsonLayer',
geojson_data,
opacity=0.2,
stroked=False,
filled=True,
extruded=True,
wireframe=True,
get_elevation='properties.pickup_count*75',
get_fill_color='[255, 255, properties.pickup_count*255/max_value]',
get_line_color=[255, 255, 255]
)

r = pdk.Deck(
layers=[polygon, geojson],
initial_view_state=INITIAL_VIEW_STATE)
return r


@classmethod
Expand Down

0 comments on commit 9b36e07

Please sign in to comment.