Criar modelos de ML com o BigQuery ML: laboratório com desafio avaliações

Task 3 doesn't work

Pierpaolo S. · Revisado há over 1 year

Md I. · Revisado há over 1 year

Ozana I. · Revisado há over 1 year

I finished Task 3 by creating and evaluating the new model improved_customer_classification_model with the required additional features, but it would never mark it as complete. Here's the query I used to create the model: CREATE OR REPLACE MODEL `ecommerce.improved_customer_classification_model` OPTIONS ( model_type='logistic_reg', labels = ['will_buy_on_return_visit'] ) AS #standardSQL SELECT * EXCEPT(fullVisitorId) FROM # features (SELECT fullVisitorId, IFNULL(totals.bounces, 0) AS bounces, IFNULL(totals.timeOnSite, 0) AS time_on_site, IFNULL(totals.pageviews, 0) AS pageviews, trafficSource.source, trafficSource.medium, channelGrouping, # mobile or desktop device.deviceCategory, # geographic IFNULL(geoNetwork.country, "") AS country FROM `data-to-insights.ecommerce.web_analytics` WHERE totals.newVisits = 1 AND date BETWEEN '20160801' AND '20170430') # train on first 9 months JOIN (SELECT fullvisitorid, IF(COUNTIF(totals.transactions > 0 AND totals.newVisits IS NULL) > 0, 1, 0) AS will_buy_on_return_visit FROM `data-to-insights.ecommerce.web_analytics` GROUP BY fullvisitorid) USING (fullVisitorId) ; And here's how I evaluated it: SELECT roc_auc, CASE WHEN roc_auc > .9 THEN 'good' WHEN roc_auc > .8 THEN 'fair' WHEN roc_auc > .7 THEN 'decent' WHEN roc_auc > .6 THEN 'not great' ELSE 'poor' END AS model_quality FROM ML.EVALUATE(MODEL ecommerce.improved_customer_classification_model, ( #standardSQL SELECT * EXCEPT(fullVisitorId) FROM # features (SELECT fullVisitorId, IFNULL(totals.bounces, 0) AS bounces, IFNULL(totals.timeOnSite, 0) AS time_on_site, IFNULL(totals.pageviews, 0) AS pageviews, trafficSource.source, trafficSource.medium, channelGrouping, # mobile or desktop device.deviceCategory, # geographic IFNULL(geoNetwork.country, "") AS country FROM `data-to-insights.ecommerce.web_analytics` WHERE totals.newVisits = 1 AND date BETWEEN '20160801' AND '20170430') # train on first 9 months JOIN (SELECT fullvisitorid, IF(COUNTIF(totals.transactions > 0 AND totals.newVisits IS NULL) > 0, 1, 0) AS will_buy_on_return_visit FROM `data-to-insights.ecommerce.web_analytics` GROUP BY fullvisitorid) USING (fullVisitorId) ));

Adam W. · Revisado há over 1 year

Kevin C. · Revisado há over 1 year

Juan Carlos B. · Revisado há over 1 year

Manish S. · Revisado há over 1 year

Jesús Arnulfo C. · Revisado há over 1 year

Ilyas N. · Revisado há over 1 year

William B. · Revisado há over 1 year

Shannon M. · Revisado há over 1 year

Shannon M. · Revisado há over 1 year

Shannon M. · Revisado há over 1 year

Andrew B. · Revisado há over 1 year

Brenda L. · Revisado há over 1 year

Dariga S. · Revisado há over 1 year

Good exercise which is a full recap for the ML Models with BigQuery ML labs serie. You need to have good SQL knowledge and remember previous exercises.

Thierry A. · Revisado há over 1 year

nartai s. · Revisado há over 1 year

Nurken S. · Revisado há over 1 year

Nurken S. · Revisado há over 1 year

Task 2 is impossible to solve. First, it's said to use "logistic_regression" then just after "use linear regression". They say to store the model in "bqml_dataset", but in Task 1 we created a dataset "bq_dataset". No way to find what you really expected to pass the test, what to trust and what to do. It has to be fixed!

Loïc B. · Revisado há over 1 year

Meghana M. · Revisado há over 1 year

Aleksandar M. · Revisado há over 1 year

Andrés Juan M. · Revisado há over 1 year

Darshan M. · Revisado há over 1 year