def main():
dc = datacommons.Client()
# Get lat/long of a city.
query = ("""
SELECT ?id ?lat ?long,
typeOf ?o City,
name ?o 'San Luis Obispo',
dcid ?o ?id,
latitude ?o ?lat,
longitude ?o ?long
""")
print('Issuing query "{}"'.format(query))
try:
df = dc.query(query)
except RuntimeError as e:
print(e)
return
with pd.option_context('display.width', 400, 'display.max_rows', 100):
print(df)
saved_file_name = dc.save_dataframe(df, 'test_df')
print(saved_file_name)
saved_df = dc.read_dataframe(saved_file_name)
assert df.equals(saved_df)
def main():
dc = datacommons.Client()
# Bootstrap with IDs of a few US cities.
pd_table = pd.DataFrame({
'city': ['City', 'dc/ve1tlm', 'dc/0vypck3', 'dc/prehdd2']
})
# Add names of those cities.
weather_table = dc.expand(pd_table, 'name', 'city', 'city_name')
with pd.option_context('display.width', 400, 'display.max_rows', 100):
print weather_table
# Add monthly mean temperature for those cities for all 2017 months.
for d in range(1, 13):
weather_table = dc.get_observations(
weather_table,
seed_col_name='city',
new_col_name=('temp_2017%.2d' % d),
start_date=('2017-%.2d-01' % d),
end_date=('2017-%.2d-01' % d),
measured_property='temperature',
stats_type='mean')
with pd.option_context('display.width', 400, 'display.max_rows', 100):
print weather_table
def main():
dc = datacommons.Client()
# Build a table with a single US state
state_table = dc.get_states('United States', 'state', max_rows=1)
# Add the state name and the 5 counties contained in that state
state_table = dc.expand(state_table,
'name',
'state',
'state name',
outgoing=True)
state_table = dc.expand(state_table,
'containedInPlace',
'state',
'county',
outgoing=False,
max_rows=3)
state_table = dc.expand(state_table,
'name',
'county',
'county name',
outgoing=True)
state_table = dc.get_populations(state_table,
seed_col_name='county',
new_col_name='county population',
population_type='Person',
max_rows=100)
with pd.option_context('display.width', 400, 'display.max_rows', 100):
print state_table
state_table = dc.get_populations(
state_table,
seed_col_name='county',
new_col_name='county_18_24_years_population',
population_type='Person',
max_rows=100,
age='USC/18To24Years')
with pd.option_context('display.width', 400, 'display.max_rows', 100):
print state_table
state_table = dc.get_populations(state_table,
seed_col_name='county',
new_col_name='county male population',
population_type='Person',
max_rows=100,
gender='Male')
with pd.option_context('display.width', 400, 'display.max_rows', 100):
print state_table
state_table = dc.get_observations(state_table,
seed_col_name='county population',
new_col_name='county person count',
observation_date='2016',
measured_property='count')
with pd.option_context('display.width', 400, 'display.max_rows', 100):
print state_table
def main():
dc = datacommons.Client()
# Get lat/long of a city.
query = ("""
SELECT ?id ?lat ?long,
typeOf ?o City,
name ?o 'San Luis Obispo',
dcid ?o ?id,
latitude ?o ?lat,
longitude ?o ?long
""")
print 'Issuing query "{}"'.format(query)
try:
df = dc.Query(query)
except RuntimeError as e:
print e
return
with pd.option_context('display.width', 400, 'display.max_rows', 100):
print df
def main():
dc = datacommons.Client()
# Start with all states in the United States and add the state names. This
# is an outgoing property of State.
pd_state = dc.get_states('United States', 'state')
pd_state = dc.expand(pd_state, 'name', 'state', 'state_name',
outgoing=True)
# Add information for counties contained in states in the 'state' column.
# Getting the county is an incoming property of State. Note that there are
# roughly 3100 counties in the United States
pd_state = dc.expand(pd_state, 'containedInPlace', 'state', 'county',
outgoing=False,
max_rows=50)
pd_state = dc.expand(pd_state, 'name', 'county', 'county_name',
outgoing=True,
max_rows=50)
# Print out the final data frame
with pd.option_context('display.width', 400, 'display.max_rows', 100):
print pd_state
def main():
dc = datacommons.Client()
# Start with all states in the United States and add the state names. This
# is an outgoing property of State.
pd_state = dc.get_places_in(
place_type='State',
container_dcid='dc/2sffw13', # United States
col_name='state')
pd_state = dc.expand(pd_state, 'name', 'state', 'state_name', outgoing=True)
# Add information for counties contained in states in the 'state' column.
# Getting the county is an incoming property of State. Note that there are
# roughly 3100 counties in the United States
pd_state = dc.expand(
pd_state,
'containedInPlace',
'state',
'county',
outgoing=False,
max_rows=50)
pd_state = dc.expand(
pd_state, 'name', 'county', 'county_name', outgoing=True, max_rows=50)
# Print out the final data frame
with pd.option_context('display.width', 400, 'display.max_rows', 100):
print pd_state
pd_city = dc.get_places_in(
place_type='City',
container_dcid='dc/b72vdv', # California
col_name='city')
pd_city = dc.expand(pd_city, 'name', 'city', 'city_name', outgoing=True)
with pd.option_context('display.width', 400, 'display.max_rows', 100):
print pd_city
def main():
dc = datacommons.Client()
# Get a list of "Class" type instance.
pd_class = dc.get_instances('class', 'Class', max_rows=_MAX_ROW)
with pd.option_context('display.width', 400, 'display.max_rows', 20):
print pd_class
# Get a list of states with their names
pd_state = dc.get_instances('state', 'State', max_rows=_MAX_ROW)
pd_state = dc.expand(pd_state,
'name',
'state',
'state_name',
outgoing=True)
with pd.option_context('display.width', 400, 'display.max_rows', 20):
print pd_state
# Get a list of cities with their names and timezone.
pd_city = dc.get_instances('city', 'City', max_rows=_MAX_ROW)
pd_city = dc.expand(pd_city, 'name', 'city', 'name', outgoing=True)
pd_city = dc.expand(pd_city, 'timezone', 'city', 'timezone', outgoing=True)
with pd.option_context('display.width', 400, 'display.max_rows', 20):
print pd_city
