def get_data(years=range(2001, 2015), replacements={}, recalculate=False):
path = os.path.join(get_data_dir("terrenos"), "terrenos_completo.csv")
if not os.path.isfile(path) or recalculate:
df = create_clean_dataset(years, replacements)
intersect_data_with_shps(
df, get_division_path("fracciones_caba_censo_2010"))
intersect_data_with_shps(df, get_division_path("radios_censo_2010"))
df.to_csv(path, encoding="utf-8")
return df
else:
df = pd.read_csv(path, encoding="utf-8")
return df.drop("Unnamed: 0", 1)
def get_data(years=range(2001, 2015), replacements={}, recalculate=False):
path = os.path.join(get_data_dir("terrenos"), "terrenos_completo.csv")
if not os.path.isfile(path) or recalculate:
df = create_clean_dataset(years, replacements)
intersect_data_with_shps(
df, get_division_path("fracciones_caba_censo_2010"))
intersect_data_with_shps(df, get_division_path("radios_censo_2010"))
df.to_csv(path, encoding="utf-8")
return df
else:
df = pd.read_csv(path, encoding="utf-8")
return df.drop("Unnamed: 0", 1)
def main(area_level="RADIO",
limit=10000,
field_name="reach_area",
transport_shp_name="recorridos_de_colectivos"):
df, old_index, new_index = get_indicators_df(area_level)
already_done = list(df[pd.notnull(df[field_name])].index)
# get bus lines
path_bus = get_transport_shp_path(transport_shp_name)
sf_lines = shapefile.Reader(path_bus)
lines = get_shapely_shapes(sf_lines)
# get division shapes
path_divisions = get_division_path(AREA_LEVEL_SHP_NAME[area_level])
sf_polys = shapefile.Reader(path_divisions)
polys = get_shapely_shapes(sf_polys)
ids = [record[0] for record in sf_polys.records()]
area_level_shapes = {
id_p: polygon
for id_p, polygon in zip(ids, polys) if id_p not in already_done
}
# iterate division shapes calculating the indicator
start = time.time()
total_shapes = len(area_level_shapes)
progress_status = "Nothing done."
for i, (id_shape, shape) in enumerate(area_level_shapes.iteritems(), 1):
# only add new value if the shape is in the index
if id_shape in new_index:
try:
lines_intersect = intersecting_lines(shape, lines)
surface = calc_reachable_surface_and_people(
shape, lines_intersect)
add_value(id_shape, surface, df)
elapsed = (time.time() - start) / 60.0
average = elapsed / i
prediction = (total_shapes - i) * average
progress_status = """
{} {}/{} in {:.2f} mins. Average: {:.2f} Time to end: {:.2f}
""".format(id_shape.ljust(10), i, total_shapes, elapsed,
average, prediction).strip()
print(progress_status, end="\r" * len(progress_status))
if i >= limit:
print(progress_status)
print("Limit of", limit, "shapes reached.")
break
except KeyboardInterrupt:
print(progress_status)
print("Interrupted!")
break
print("Saving values...", end=" ")
save_indicators_df(df, old_index)
print("Done.")
def main(area_level="RADIO", limit=10000, field_name="reach_area",
transport_shp_name="recorridos_de_colectivos"):
df, old_index, new_index = get_indicators_df(area_level)
already_done = list(df[pd.notnull(df[field_name])].index)
# get bus lines
path_bus = get_transport_shp_path(transport_shp_name)
sf_lines = shapefile.Reader(path_bus)
lines = get_shapely_shapes(sf_lines)
# get division shapes
path_divisions = get_division_path(AREA_LEVEL_SHP_NAME[area_level])
sf_polys = shapefile.Reader(path_divisions)
polys = get_shapely_shapes(sf_polys)
ids = [record[0] for record in sf_polys.records()]
area_level_shapes = {id_p: polygon for id_p, polygon in zip(ids, polys) if
id_p not in already_done}
# iterate division shapes calculating the indicator
start = time.time()
total_shapes = len(area_level_shapes)
progress_status = "Nothing done."
for i, (id_shape, shape) in enumerate(area_level_shapes.iteritems(), 1):
# only add new value if the shape is in the index
if id_shape in new_index:
try:
lines_intersect = intersecting_lines(shape, lines)
surface = calc_reachable_surface_and_people(shape,
lines_intersect)
add_value(id_shape, surface, df)
elapsed = (time.time() - start) / 60.0
average = elapsed / i
prediction = (total_shapes - i) * average
progress_status = """
{} {}/{} in {:.2f} mins. Average: {:.2f} Time to end: {:.2f}
""".format(id_shape.ljust(10), i, total_shapes, elapsed,
average, prediction).strip()
print(progress_status, end="\r" * len(progress_status))
if i >= limit:
print(progress_status)
print("Limit of", limit, "shapes reached.")
break
except KeyboardInterrupt:
print(progress_status)
print("Interrupted!")
break
print("Saving values...", end=" ")
save_indicators_df(df, old_index)
print("Done.")
def find_containing_radios(df):
shp_path = get_division_path("radios_censo_2010")
sf = shapefile.Reader(shp_path)
radios_dict = {int(comuna): {} for comuna in df["comuna"].unique()}
for sr in sf.iterShapeRecords():
shape = get_shapely_shape(sr.shape)
radio = sr.record[0]
comuna = int(radio.split("_")[0])
radios_dict[comuna][radio] = shape
return df.apply(find_containing_shape, axis=1,
args=(radios_dict, shp_path))
def find_containing_radios(df):
shp_path = get_division_path("radios_censo_2010")
sf = shapefile.Reader(shp_path)
radios_dict = {int(comuna): {} for comuna in df["comuna"].unique()}
for sr in sf.iterShapeRecords():
shape = get_shapely_shape(sr.shape)
radio = sr.record[0]
comuna = int(radio.split("_")[0])
radios_dict[comuna][radio] = shape
return df.apply(find_containing_shape,
axis=1,
args=(radios_dict, shp_path))
def _prepare_test_case():
path_colectivos = get_transport_shp_path("recorrido-colectivos")
path_radios = get_division_path("radios_censo_2010")
sf_lines = shapefile.Reader(path_colectivos)
lines = get_shapely_shapes(sf_lines)
sf_polys = shapefile.Reader(path_radios)
polys = get_shapely_shapes(sf_polys)
ids = [record[0] for record in sf_polys.records()]
radios = {id: polygon for id, polygon in zip(ids, polys)}
radio = radios["14_1_8"]
return radio, lines