joshreini1

tru.set_environment("remote")
tru.add_project("BQML Integration Example", score_type = "logits")

joshreini1

query_attributions = """#standardsql
SELECT
  (SELECT attribution FROM UNNEST(top_feature_attributions) WHERE feature = 'pageviews') AS pageviews,
  (SELECT attribution FROM UNNEST(top_feature_attributions) WHERE feature = 'country') AS country,
  (SELECT attribution FROM UNNEST(top_feature_attributions) WHERE feature = 'is_mobile') AS device_id,
  (SELECT attribution FROM UNNEST(top_feature_attributions) WHERE feature = 'os') AS os
FROM 'gcp-sandbox-311420.bqml_tutorial.sample model preds explanations 0701 0708'
"""
job_attr = client.query(query_attributions)
df_attr = job_attr.to_dataframe()

joshreini1

from google.cloud.bigquery import Client, QueryJobConfig
client = Client()

joshreini1

!pip install truera

import truera
from truera.client.truera_workspace import TrueraWorkspace
from truera.client.truera_authentication import TokenAuthentication

tru = TrueraWorkspace('https://app.truera.net/', TokenAuthentication(TOKEN))

joshreini1

#standardSQL
CREATE MODEL 'bqml_tutorial.sample_model'
OPTIONS(model_type='boosted_tree_classifier') AS
SELECT
  IF(totals.transactions IS NULL, 0, 1) AS label,
  IFNULL(device.operatingSystem, "") as os,
  device.isMobile AS is_mobile,
  IFNULL(geoNetwork.country, "") AS country,
  IFNULL(totals.pageviews, 0) AS pageviews
FROM

joshreini1

#standardSQL
SELECT
  visitorID, #passthrough
  Country,
  Top_feature_attributions, # from ml.explain_predict
  SUM(predicted_label) as total_predicted_purchases,
  label #passthrough
FROM
  ML.EXPLAIN_PREDICT(MODEL `bqml_tutorial.sample_model`, (
SELECT

joshreini1

! pip install truera
from truera.client.truera_workspace import TrueraWorkspace #import truera
from truera.client.truera_authentication import TokenAuthentication # import authentication
tru = TrueraWorkspace("https://app.truera.net", TokenAuthentication("ADD YOUR AUTH TOKEN")) 
tru.add_project("My project", score_type="CHOOSE A SCORE TYPE: regression, classification, probits or logits")
tru.add_data_collection("data_collection_1") # the schema that will hold your data "splits" and model
tru.add_data(
    data = data,
    data_split_name = "train",
    column_spec=ColumnSpec(

joshreini1

def rebalance_gender(df, data_type):
   if data_type == 0:
       df_female_true = df[(df['Sex'] == 'Female') & (df['PINCP'] == True)]
       df_else = df[~((df['Sex'] == 'Female') & (df['PINCP'] == True))]
   else:
       df_female_true = df[(df['Sex_Female'] == 1) & (df['PINCP'] == True)]
       df_else = df[~((df['Sex_Female'] == 1) & (df['PINCP'] == True))]


   if data_type == 0:

joshreini1

# add data splits to the collection we just created
year_begin = 2014
year_end = 2016 # exclusive
states = ['CA', 'NY']


for year in range(year_begin, year_end):
   for state in states:
       tru.add_data_split(f'{year}-{state}',
         pre_data = data[year][state]['data_preprocessed'].\

joshreini1

feature_map.pop('Sex')
tru.add_data_collection("Data Collection v2",
                       pre_to_post_feature_map = feature_map,
                       provide_transform_with_model = False)