DataOps

Generalizing the scikit-learn pipeline. See skrub DataOps for further details.

var	Create a skrub variable.
X	Create a skrub variable and mark it as being X.
y	Create a skrub variable and mark it as being y.
as_data_op	Create a DataOp DataOp that evaluates to the given value.
deferred	Wrap function calls in a DataOp DataOp.

The DataOp object.

DataOp

Representation of a computation that can be used to build DataOps plans and learners.

Inline hyperparameters selection in your DataOps plan.

choose_bool	A choice between True and False.
choose_float	A choice of floating-point numbers from a numeric range.
choose_int	A choice of integers from a numeric range.
choose_from	A choice among several possible outcomes.
optional	A choice between value and None.

Evaluate your DataOps plan.

cross_validate	Cross-validate a learner built from a DataOp.
eval_mode	Return the mode in which the DataOp is currently being evaluated.

The skb accessor exposes all DataOps methods and attributes.

DataOp.skb.apply	Apply a scikit-learn estimator to a dataframe or numpy array.
DataOp.skb.apply_func	Apply the given function.
DataOp.skb.clone	Get an independent clone of the DataOp.
DataOp.skb.concat	Concatenate dataframes vertically or horizontally.
DataOp.skb.cross_validate	Cross-validate the DataOp plan.
DataOp.skb.describe_defaults	Describe the hyper-parameters used by the default learner.
DataOp.skb.describe_param_grid	Describe the hyper-parameters extracted from choices in the DataOp.
DataOp.skb.describe_steps	Get a text representation of the computation graph.
DataOp.skb.draw_graph	Get an SVG string representing the computation graph.
DataOp.skb.drop	Drop some columns.
DataOp.skb.eval	Evaluate the DataOp.
DataOp.skb.freeze_after_fit	Freeze the result during learner fitting.
DataOp.skb.full_report	Generate a full report of the DataOp's evaluation.
DataOp.skb.get_data	Collect the values of the variables contained in the DataOp.
DataOp.skb.make_learner	Get a skrub learner for this DataOp.
DataOp.skb.make_grid_search	Find the best parameters with grid search.
DataOp.skb.make_randomized_search	Find the best parameters with randomized search.
DataOp.skb.if_else	Create a conditional DataOp.
DataOp.skb.iter_learners_grid	Get learners with different parameter combinations.
DataOp.skb.iter_learners_randomized	Get learners with different parameter combinations.
DataOp.skb.mark_as_X	Mark this DataOp as being the X table.
DataOp.skb.mark_as_y	Mark this DataOp as being the y table.
DataOp.skb.match	Select based on the value of a DataOp.
DataOp.skb.preview	Get the value computed for previews (shown when printing the DataOp).
DataOp.skb.select	Select a subset of columns.
DataOp.skb.set_description	Give a description to this DataOp.
DataOp.skb.set_name	Give a name to this DataOp.
DataOp.skb.subsample	Configure subsampling of a dataframe or numpy array.
DataOp.skb.train_test_split	Split an environment into a training an testing environments.

Accessor attributes.

DataOp.skb.description	A user-defined description or comment about the DataOp.
DataOp.skb.is_X	Whether this DataOp has been marked with skb.mark_as_X().
DataOp.skb.is_y	Whether this DataOp has been marked with skb.mark_as_y().
DataOp.skb.name	A user-chosen name for the DataOp.
DataOp.skb.applied_estimator	Retrieve the estimator applied in the previous step, as a DataOp.

Objects generated by the DataOps.

SkrubLearner	Learner that evaluates a skrub DataOp.
ParamSearch	Learner that evaluates a skrub DataOp with hyperparameter tuning.