Logistic Regression

SQL function: cuml_logistic_regression

Official cuML reference: Python API

Fit a query-local cuML logistic or softmax classifier and score independent rows.

Quickstart

Register the relations referenced by the parenthesized SELECT clauses below. For metadata validation, the descriptor names training_vectors and predict_vectors.

SELECT id, prediction FROM cuml_logistic_regression(training => (SELECT id, d0, label FROM train), predict => (SELECT id, d0 FROM test)) ORDER BY row_ordinal;

Inputs

Each relation argument is a parenthesized SELECT subquery. Metadata validation resolves a registered table or view for the same role without scanning its rows. See ML Inputs for the ID, Float32 feature, null, finite-value, and runtime-dimension contract.

Role	Required	Validation reference	Description
training	yes	table	Dense Float32 features and non-null Int32 label used to fit the query-local model.
predict	yes	table	Independent dense Float32 rows scored with the query-local model; must omit label.

Vector element types

Element type	Valid metrics
Float32	Not applicable

Arguments and options

Scalar SQL arguments

Argument	Type	Required	Description
penalty_l1	number	no	L1 penalty.
penalty_l2	number	no	L2 penalty.
max_iter	integer	no	Maximum QN iterations.
linesearch_max_iter	integer	no	Maximum line-search iterations.
lbfgs_memory	integer	no	L-BFGS history length.
fit_intercept	boolean	no	Fit an intercept.
penalty_normalized	boolean	no	Normalize the penalty by row count.
grad_tol	number	no	QN gradient tolerance.
change_tol	number	no	QN objective change tolerance.

SQL value argument schemas

Argument	Required	Literal shape	Default	Constraints	Description
change_tol	no	number	0.00001	minimum 0	QN objective change tolerance.
fit_intercept	no	boolean	true		Fit an intercept.
grad_tol	no	number	0.0001	minimum 0	QN gradient tolerance.
lbfgs_memory	no	integer	5	minimum 1; maximum 2147483647	L-BFGS history length.
linesearch_max_iter	no	integer	50	minimum 1; maximum 2147483647	Maximum line-search iterations.
max_iter	no	integer	1000	minimum 1; maximum 2147483647	Maximum QN iterations.
penalty_l1	no	number	0	minimum 0	L1 penalty.
penalty_l2	no	number	0	minimum 0	L2 penalty.
penalty_normalized	no	boolean	true		Normalize the penalty by row count.

Vector binding shapes

Each relation subquery projects a non-null id and either non-null Float32 feature columns in dimension order or one non-null vector list column of non-null Float32 values. For classification, training also projects non-null Int32 label; regression requires non-null finite Float32 label. The predict relation omits it.

The list-column shape excludes other feature columns. See ML Inputs for the allowed list containers and runtime checks.

Output

Column	Type	Nullable	Description
row_ordinal	UInt64	no	Evaluated input row position, including when IDs repeat.
id	same_as_predict.id	no	Logical ID from the predict relation.
prediction	Int32	no	Predicted zero-based class index.

Concrete schemas are call-specific. Run gpu_validate_call against registered relations to inspect the output schema after the actual ID types and literal options are validated.

Limits

• Validation resolves named tables or views and reads schemas only; it does not execute relation scans or GPU work.
• Execution relation arguments require parenthesized subqueries; dry-run validation accepts registered named relations only.
• Training labels must be non-null Int32 contiguous classes 0..C-1; predict must omit label and match the feature dimension.
• Binary labels use logistic loss; three or more classes use softmax. Reaching max_iter is not an error.
• The model is query-local. An empty predict relation returns an empty result without a native call.

To dry-run validate relation metadata, column types, and options without execution, see gpu_validate_call.