def get_home_locations_by_census_tract(siteid, state_code, county_code):
census_tract_geographies = get_from_data_source('CENSUS_TRACT_DF')
census_tract_geographies = census_tract_geographies[
(census_tract_geographies['state_code'] == state_code)
& (census_tract_geographies['county_code'] == county_code)]
home_locations = get_from_data_source('HOME_LOCATIONS_DF')
site_home_locations = home_locations[
(home_locations['siteid'] == siteid)
& (home_locations['state_code'] == state_code)
& (home_locations['county_code'] == county_code)]
site_home_locations = site_home_locations[[
'tract', 'visit_days', 'visitors_unq'
]]
site_home_census_data = census_tract_geographies.merge(site_home_locations,
on='tract',
how='inner')
svi_df = get_from_data_source('SVI_DF')
svi_df = svi_df[svi_df['state_code'] == state_code]
svi_df = svi_df.drop(columns=['state_code'])
site_home_census_data = site_home_census_data.merge(svi_df,
on='tract',
how='inner')
return site_home_census_data
def get_project_home_locations_by_census_tract(project, state_code,
county_code):
census_tract_geographies = get_from_data_source('CENSUS_TRACT_DF')
census_tract_geographies = census_tract_geographies[
(census_tract_geographies['state_code'] == state_code)
& (census_tract_geographies['county_code'] == county_code)]
project_sites = get_project_sites(project)
project_site_ids = project_sites['siteid'].drop_duplicates()
home_locations = get_from_data_source('HOME_LOCATIONS_DF')
project_home_locations = home_locations[
(home_locations['siteid'].isin(project_site_ids))
& (home_locations['state_code'] == state_code)
& (home_locations['county_code'] == county_code)]
# for each tract, we need to sum visit days and visitors_unq
project_home_locations = project_home_locations.groupby(
by=['tract'], as_index=False)['visit_days', 'visitors_unq'].sum()
project_home_census_data = census_tract_geographies.merge(
project_home_locations, on='tract', how='inner')
svi_df = get_from_data_source('SVI_DF')
svi_df = svi_df[svi_df['state_code'] == state_code]
svi_df = svi_df.drop(columns=['state_code'])
project_home_census_data = project_home_census_data.merge(svi_df,
on='tract',
how='inner')
return project_home_census_data
def get_home_locations_by_state(siteid):
state_boundaries = get_from_data_source('STATE_BOUNDARIES_DF')
home_locations = get_from_data_source('HOME_LOCATIONS_DF')
site_home_locations = home_locations[home_locations['siteid'] == siteid]
site_home_locations = site_home_locations[[
'state_code', 'state', 'visit_days', 'visitors_unq'
]]
site_home_locations = site_home_locations.groupby(
by=['state_code', 'state'], as_index=False).sum()
site_home_state_data = state_boundaries.merge(site_home_locations,
on=['state_code', 'state'],
how='inner')
return site_home_state_data
def get_demographic_summary(siteid):
svi_df = get_from_data_source('SVI_DF')
home_locations = get_from_data_source('HOME_LOCATIONS_DF')
project = get_project_from_site(siteid)
census_tract_states = app_config.CENSUS_TRACT_STATES[project]
svi_df = svi_df[svi_df['state_code'].isin(census_tract_states)]
svi_df = svi_df.drop(columns=['state_code'])
site_home_locations = home_locations[home_locations['siteid'] == siteid]
demographics_data = site_home_locations.merge(svi_df,
on='tract',
how='inner')
return demographics_data
def _get_project_estimates(project, period):
project_sites = get_project_sites(project)
project_site_ids = project_sites['siteid'].drop_duplicates()
df = get_from_data_source('MONTHLY_VISITATION_DF')
project_sites_data = df[df['trail'].isin(project_site_ids)]
if period == 'monthly':
project_sites_data = project_sites_data.drop(columns='year').groupby(
by=['trail', 'month'], as_index=False).mean()
project_sites_data = project_sites_data.drop(columns='trail').groupby(
by=['month']).sum()
elif period == 'annual':
project_sites_data = project_sites_data.drop(
columns=['trail', 'month']).groupby(by=['year']).sum()
project_sites_data = project_sites_data[[
'estimate', 'log_estimate', 'flickr', 'twitter', 'instag', 'wta',
'alltrails', 'onsite', 'log_onsite', 'data_days'
]]
project_sites_data['log_estimate'] = np.log(
project_sites_data['estimate'] + 1)
project_sites_data['log_onsite'] = np.log(project_sites_data['onsite'] + 1)
project_sites_data.reset_index(drop=False, inplace=True)
return project_sites_data
def get_project_readme(readme_type, project=None):
project_readme_cache = get_from_data_source('PROJECT_README')
if readme_type == 'VISITS':
return project_readme_cache[project + '_VISITS']
elif readme_type == 'INFO':
return project_readme_cache[project]
else:
return project_readme_cache[readme_type]
def get_project_home_locations_by_state(project):
state_boundaries = get_from_data_source('STATE_BOUNDARIES_DF')
home_locations = get_from_data_source('HOME_LOCATIONS_DF')
project_sites = get_project_sites(project)
project_site_ids = project_sites['siteid'].drop_duplicates()
project_home_locations = home_locations[home_locations['siteid'].isin(
project_site_ids)]
project_home_locations = project_home_locations[[
'state_code', 'state', 'visit_days', 'visitors_unq'
]]
project_home_locations = project_home_locations.groupby(
by=['state_code', 'state'], as_index=False).sum()
project_home_state_data = state_boundaries.merge(
project_home_locations, on=['state_code', 'state'], how='inner')
return project_home_state_data
def _get_project_visitation_data(project, period):
project_sites = get_project_sites(project)
project_site_ids = project_sites['siteid'].drop_duplicates()
if period == 'monthly':
df = get_from_data_source('MONTHLY_VISITATION_DF')
group_by_cols = ['year', 'month']
else:
df = get_from_data_source('WEEKLY_VISITATION_DF')
group_by_cols = ['year', 'month', 'week']
project_sites_data = df[df['trail'].isin(project_site_ids)]
project_sites_data = project_sites_data.drop('trail', axis=1)
project_sites_data = project_sites_data.groupby(group_by_cols,
as_index=False).sum()
project_sites_data['log_estimate'] = np.log(
project_sites_data['estimate'] + 1)
return project_sites_data
def get_home_locations_by_county(siteid, state_code):
county_geographies = get_from_data_source('COUNTIES_DF')
county_geographies = county_geographies[county_geographies['state_code'] ==
state_code]
home_locations = get_from_data_source('HOME_LOCATIONS_DF')
site_home_locations = home_locations[
(home_locations['siteid'] == siteid)
& (home_locations['state_code'] == state_code)]
site_home_locations = site_home_locations[[
'county_code', 'county', 'visit_days', 'visitors_unq'
]]
site_home_locations = site_home_locations.groupby(
by=['county_code', 'county'], as_index=False).sum()
site_home_county_data = county_geographies.merge(
site_home_locations, on=['county_code', 'county'], how='inner')
return site_home_county_data
def get_project_demographic_summary(project):
svi_df = get_from_data_source('SVI_DF')
home_locations = get_from_data_source('HOME_LOCATIONS_DF')
project_sites = get_project_sites(project)
project_site_ids = project_sites['siteid'].drop_duplicates()
project_home_locations = home_locations[home_locations['siteid'].isin(
project_site_ids)]
project_home_locations = project_home_locations.groupby(
by=['tract'], as_index=False).sum()
census_tract_states = app_config.CENSUS_TRACT_STATES[project]
svi_df = svi_df[svi_df['state_code'].isin(census_tract_states)]
svi_df = svi_df.drop(columns=['state_code'])
demographics_data = project_home_locations.merge(svi_df,
on='tract',
how='inner')
return demographics_data
def get_project_last_year_estimates(project):
project_sites = get_project_sites(project)
project_site_ids = project_sites['siteid'].drop_duplicates()
df = get_from_data_source('MONTHLY_VISITATION_DF')
project_sites_data = df[df['trail'].isin(project_site_ids)]
project_sites_data = project_sites_data[project_sites_data['year'] == 2018]
project_sites_data = project_sites_data.groupby(by=['trail'],
as_index=False).sum()
project_sites_data = project_sites_data[['trail', 'estimate']]
return project_sites_data
def _get_estimates(siteid, period):
monthly_df = get_from_data_source('MONTHLY_VISITATION_DF')
site_data = monthly_df[monthly_df['trail'] == siteid]
site_data = site_data.groupby(
by=['month']).mean() if period == 'monthly' else site_data.groupby(
by=['year']).sum()
site_data = site_data[[
'estimate', 'log_estimate', 'flickr', 'twitter', 'instag', 'wta',
'alltrails', 'ebird', 'onsite', 'log_onsite', 'data_days'
]]
site_data.reset_index(inplace=True)
return site_data
def get_project_home_locations_by_county(project, state_code):
county_geographies = get_from_data_source('COUNTIES_DF')
county_geographies = county_geographies[county_geographies['state_code'] ==
state_code]
home_locations = get_from_data_source('HOME_LOCATIONS_DF')
project_sites = get_project_sites(project)
project_site_ids = project_sites['siteid'].drop_duplicates()
project_home_locations = home_locations[
(home_locations['siteid'].isin(project_site_ids))
& (home_locations['state_code'] == state_code)]
project_home_locations = project_home_locations[[
'county_code', 'county', 'visit_days', 'visitors_unq'
]]
project_home_locations = project_home_locations.groupby(
by=['county_code', 'county'], as_index=False).sum()
project_home_county_data = county_geographies.merge(
project_home_locations, on=['county_code', 'county'], how='inner')
return project_home_county_data
def get_project_home_locations(project):
home_locations = get_from_data_source('HOME_LOCATIONS_DF')
project_sites = get_project_sites(project)
project_site_ids = project_sites['siteid'].drop_duplicates()
project_home_locations = home_locations[home_locations['siteid'].isin(
project_site_ids)]
# International doesn't have state and counties and hence a three column group by will exclude it
# extra effort to include international
international_visits = project_home_locations[
project_home_locations['country'] == 'International']
international_visits = international_visits.groupby(['country'],
as_index=False).sum()
project_home_locations = project_home_locations.groupby(
by=['country', 'state', 'county'], as_index=False).sum()
project_home_locations = project_home_locations.append(
international_visits, ignore_index=True, sort=False)
return _treefy_home_locations(project, project_home_locations)
def get_project_sites(project_group):
allsites = get_from_data_source('ALLSITES_DF')
project_sites = allsites[allsites['Prjct_code'].str.contains(
project_group)]
return project_sites
def get_monthly_visitation(siteid):
monthly_df = get_from_data_source('MONTHLY_VISITATION_DF')
site_data = monthly_df[monthly_df['trail'] == siteid]
return site_data
def get_project_from_site(siteid):
allsites = get_from_data_source('ALLSITES_DF')
site = allsites[allsites['siteid'] == siteid]
return site[['Prjct_code']].iat[0, 0]
def get_home_locations(siteid):
home_locations = get_from_data_source('HOME_LOCATIONS_DF')
site_home_locations = home_locations[home_locations['siteid'] == siteid]
return _treefy_home_locations(siteid, site_home_locations)
def get_weekly_visitation(siteid):
weekly_df = get_from_data_source('WEEKLY_VISITATION_DF')
site_data = weekly_df[weekly_df['trail'] == siteid]
return site_data