bigquery ml - bobbae/gcp GitHub Wiki

BigQuery ML enables users to create and execute machine learning models in BigQuery by using SQL queries.

https://towardsdatascience.com/creating-a-machine-learning-model-with-sql-81d843a5307c

https://medium.com/@alidoggaz/doing-ai-with-sql-syntax-bigquery-cea910b47bb7

BigQuery ML Algorithms supported

k-means clustering is a method of vector quantization, originally from signal processing, that aims to partition n observations into k clusters in which each observation belongs to the cluster with the nearest mean (cluster centers or cluster centroid).

https://en.wikipedia.org/wiki/K-means_clustering

Multiclass logistic regression

For classification scenarios when the choice is between multiple options

https://gluon.mxnet.io/chapter02_supervised-learning/softmax-regression-scratch.html

Matrix factorization

https://cloud.google.com/bigquery-ml/docs/bigqueryml-mf-explicit-tutorial

For developing recommendation engines based on past information

https://developers.google.com/machine-learning/recommendation/collaborative/matrix

Time series and ARIMA

https://en.wikipedia.org/wiki/Autoregressive_integrated_moving_average

https://cloud.google.com/bigquery-ml/docs/arima-single-time-series-forecasting-tutorial

To forecast business KPIs leveraging time series data from the past

https://otexts.com/fpp2/arima.html

Boosted tree

https://xgboost.readthedocs.io/en/latest/tutorials/model.html

For classification and regression use cases with XGBoost

https://github.com/PacktPublishing/Machine-Learning-with-BigQuery-ML/blob/master/Chapter10/classifying_trees_xgboost.sql

AutoML Tables

To leverage AutoML capabilities from the BigQuery SQL interface

https://cloud.google.com/automl-tables/docs

Deep Neural Network (DNN)

For developing TensorFlow models for classification or regression scenarios, avoiding any lines of code

https://machinelearningmastery.com/what-is-deep-learning/

BigQuery ML NYC Bike Sharing DNN Example

https://github.com/PacktPublishing/Machine-Learning-with-BigQuery-ML/blob/master/Chapter11/nyc_bike_sharing_dnn.sql

Explainable AI and BigQuery ML

https://medium.com/codex/using-explainable-ai-in-bigquery-ml-1ec9ac0a7228

Vertex AI Pipelines and BigQuery ML

https://cloud.google.com/blog/topics/developers-practitioners/announcing-bigquery-and-bigquery-ml-operators-vertex-ai-pipelines

BigQuery ML Tutorials

https://cloud.google.com/bigquery-ml/docs/tutorials

Using AutoML Tables with BigQuery ML

AutoML Tables let you build and deploy machine learning models using structured data and explain prediction results. It can be used along with BigQuery ML. This makes it easy to train Tables models on BigQuery data using SQL from BigQuery UI and evaluate and use the models for prediction directly via SQL.

https://amygdala.github.io/gcp_blog/ml/automl/bigquery/2020/07/14/bqml_tables.html

To train an AutoML Tables model inside of BigQuery, use the BigQuery ML CREATE MODEL statement with one of the AutoML Tables model types.

https://cloud.google.com/bigquery-ml/docs/reference/standard-sql/bigqueryml-syntax-create-automl