What problem does it solve? Writing pandas-style data analysis and machine learning code against BigQuery requires knowing BigFrames-specific APIs, ordering modes, and ML packages that differ from standard pandas and scikit-learn, and mistakes cause slow queries or out-of-memory errors. ## Core Features & Use Cases - BigFrames DataFrame Best Practices: Enforces partial ordering mode, peek() previews, accessor-based transformations, and avoidance of local materialization via to_pandas(). - Serverless ML with BigQuery: Guides model training, evaluation, and prediction through bigframes.bigquery.ml instead of scikit-learn, with reference examples for linear and logistic regression. - Legacy BigFrames ML Support: Covers the legacy bigframes.ml package including ARIMA Plus forecasting, PCA, and model persistence with to_gbq(). - Use Case: In a notebook, ask for a logistic regression model trained on the public penguins dataset, and receive correct BigFrames code that trains directly in BigQuery without downloading data. ## Quick Start Write BigFrames code that loads the bigquery-public-data penguins table and trains a linear regression model to predict penguin body mass.
