Tutorials

All BigQuery ML tutorials

• Creating a regression model to predict penguin weight

  This tutorial uses a linear regression model in BigQuery ML to predict the weight of a penguin.

• Creating a classification model on census data

  This tutorial uses a binary logistic regression model in BigQuery ML to predict the income range of respondents in the US Census Dataset

• Creating a k-means model to cluster London bicycle hires dataset

  This tutorial uses a k-means model in BigQuery ML to identify clusters of data in the London Bicycle Hires public dataset.

• Creating a matrix factorization model to make movie recommendations

  This tutorial uses the public movielens dataset to create a model from explicit feedback that generates movie recommendations for a user.

• Creating a matrix factorization model to make recommendations from Google Analytics data

  This tutorial uses the BigQuery Analytics sample to create a model from implicit feedback that recommends content for a visitor to a website.

• Single time-series forecasting from Analytics data

  This tutorial creates a time series model to perform single time-series forecasts using the google_analytics_sample.ga_sessions sample table.

• Multiple time-series forecasting with a single query for NYC Citi Bike trips

  This tutorial creates a set of time-series models to perform multiple time-series forecasts with a single query. You will use the new_york.citibike_trips data. This data contains information about Citi Bike trips in New York City.

• Scalable forecasting with millions of time-series in BigQuery

  This tutorial uses a set of techniques to enable 100x faster forecasting without sacrificing much forecasting accuracy. It enables forecasting millions of time series within hours using a single query.

• Multivariate time-series forecasting from Seattle air quality data

  This tutorial shows how to create a multivariate time series model to perform time-series forecasting. This enables forecasting with extra features.

• Using `TRANSFORM` clause for feature engineering

  This tutorial uses the BigQuery ML `TRANSFORM` clause for feature engineering to create a model that predicts the birth weight of a child.

• Using hyperparameter tuning to improve model performance

  This tutorial uses the tlc_yellow_trips_2018 sample table to create a model with hyperparameter tuning that predicts the tip of a taxi trip.

• Importing TensorFlow models to make predictions

  This tutorial imports a TensorFlow model into a BigQuery dataset and use it to make predictions from a SQL query.

• Exporting a BigQuery ML model for online prediction

  This tutorial exports a BigQuery ML model and then deploys the model either on AI Platform or on a local machine. You will use the iris table from the BigQuery public datasets.