janbenetka

SELECT * FROM subset 
WHERE RAND() < 1e6 / (SELECT COUNT(*) FROM subset)

janbenetka

from IPython.display import HTML, IFrame, display
display(HTML("<style>.container { width:1400px !important; }</style>"))

janbenetka

PREFIX dbpedia2: <http://dbpedia.org/property/> 
PREFIX dbo: <http://dbpedia.org/ontology/>
PREFIX geo: <http://www.w3.org/2003/01/geo/wgs84_pos#>

SELECT *
WHERE {
  ?county rdf:type yago:County108546183 .
  ?county dbo:country dbr:United_States.
  ?county dbo:state ?state.
  ?county rdfs:label ?county_label.

janbenetka

CREATE OR REPLACE TABLE `uc-prox-core-dev.z_jan_retained.adjacent_tracts` AS

WITH adjacent_tracts AS (
  SELECT a.fips, b.fips as fips_adjacent, a.state_name, b.state_name as state_name_adjacent, a.county_name, b.county_name as county_name_adjacent, a.geog, b.geog as geog_adjacent,
  FROM `uc-atlas.maps_us.census_tracts` a, `uc-atlas.maps_us.census_tracts` b
  WHERE ST_INTERSECTS(ST_BUFFER(a.geog, 25), b.geog)
)

SELECT * FROM adjacent_tracts

janbenetka

from IPython.display import Javascript  # Restrict height of output cell.
display(Javascript('''google.colab.output.setIframeHeight(0, true, {maxHeight: 300})'''))

janbenetka

WITH osm AS (
  SELECT 
  layer_code, 
  layer_class, 
  layer_name, 
  gdal_type as geography_type, 
  osm_id, 
  CASE 
    WHEN (tags.key = 'name') THEN tags.value
    ELSE ''

janbenetka

WITH daily AS (
  SELECT 
    local_event_date, 
    venue_type,    
    SUM(device_count) device_count, 
    SUM(person_count) person_count,
  FROM `uc-prox-core-dev.international_metrics.traffic_trends` 
  GROUP BY local_event_date, venue_type
)

janbenetka

dict_filter = lambda x, y: dict([ (i,x[i]) for i in x if i in set(y) ])
large_dict = {"a":1,"b":2,"c":3,"d":4}
keys_to_remove = ("c","d")
updated_dict = dict_filter(large_dict, keys_to_remove)
print(updated_dict)

janbenetka

import importlib
importlib.reload(scale_by_scale_optim)

from utils import scale_by_scale_optim

janbenetka

def get_accuracy(series, recovered_series):
    """Compare a simulated and recovered series in terms of overall label accuracy.
    
    Input
    -----
        series : list of lists
        recovered_series : list of lists
        
    Output
    ------