An alias for getOrDefault().

Param for block size for stacking input data in matrices. Data is stacked within partitions. If block size is more than remaining data in a partition then it is adjusted to the size of this data..

Clears the user-supplied value for the input param.

Creates a copy of this instance with the same UID and some extra params. Subclasses should implement this method and set the return type properly. See defaultCopy().

Copies param values from this instance to another instance for params shared by them.

This handles default Params and explicitly set Params separately. Default Params are copied from and to defaultParamMap, and explicitly set Params are copied from and to paramMap. Warning: This implicitly assumes that this Params instance and the target instance share the same set of default Params.

Default implementation of copy with extra params. It tries to create a new instance with the same UID. Then it copies the embedded and extra parameters over and returns the new instance.

Explains a param.

Explains all params of this instance. See explainParam().

extractParamMap with no extra values.

Extracts the embedded default param values and user-supplied values, and then merges them with extra values from input into a flat param map, where the latter value is used if there exist conflicts, i.e., with ordering: default param values less than user-supplied values less than extra.

Param for features column name.

Fits a model to the input data.

Fits multiple models to the input data with multiple sets of parameters. The default implementation uses a for loop on each parameter map. Subclasses could override this to optimize multi-model training.

Fits a single model to the input data with provided parameter map.

Fits a single model to the input data with optional parameters.

Optionally returns the user-supplied value of a param.

Gets the default value of a parameter.

Get the number of classes. This looks in column metadata first, and if that is missing, then this assumes classes are indexed 0,1,...,numClasses-1 and computes numClasses by finding the maximum label value.

Label validation (ensuring all labels are integers >= 0) needs to be handled elsewhere, such as in extractLabeledPoints().

Gets the value of a param in the embedded param map or its default value. Throws an exception if neither is set.

Gets a param by its name.

Tests whether the input param has a default value set.

Tests whether this instance contains a param with a given name.

The initial weights of the model.

Checks whether a param is explicitly set or has a default value.

Checks whether a param is explicitly set.

Param for label column name.

Layer sizes including input size and output size.

Param for maximum number of iterations (>= 0).

Returns all params sorted by their names. The default implementation uses Java reflection to list all public methods that have no arguments and return Param.

Param for prediction column name.

Param for Column name for predicted class conditional probabilities. Note: Not all models output well-calibrated probability estimates! These probabilities should be treated as confidences, not precise probabilities.

Param for raw prediction (a.k.a. confidence) column name.

Saves this ML instance to the input path, a shortcut of write.save(path).

Param for random seed.

Sets a parameter in the embedded param map.

Sets a parameter (by name) in the embedded param map.

Sets a parameter in the embedded param map.

Sets the value of param blockSize. Default is 128.

Sets default values for a list of params.

Note: Java developers should use the single-parameter setDefault. Annotating this with varargs can cause compilation failures due to a Scala compiler bug. See SPARK-9268.

Sets a default value for a param.

Sets the value of param initialWeights.

Sets the value of param layers.

Set the maximum number of iterations. Default is 100.

Set the seed for weights initialization if weights are not set

Sets the value of param solver. Default is "l-bfgs".

Sets the value of param stepSize (applicable only for solver "gd"). Default is 0.03.

Set the convergence tolerance of iterations. Smaller value will lead to higher accuracy with the cost of more iterations. Default is 1E-6.

The solver algorithm for optimization. Supported options: "gd" (minibatch gradient descent) or "l-bfgs". Default: "l-bfgs"

Param for Step size to be used for each iteration of optimization (> 0).

Param for Thresholds in multi-class classification to adjust the probability of predicting each class. Array must have length equal to the number of classes, with values > 0 excepting that at most one value may be 0. The class with largest value p/t is predicted, where p is the original probability of that class and t is the class's threshold.

Param for the convergence tolerance for iterative algorithms (>= 0).

Train a model using the given dataset and parameters. Developers can implement this instead of fit() to avoid dealing with schema validation and copying parameters into the model.

Check transform validity and derive the output schema from the input schema.

We check validity for interactions between parameters during transformSchema and raise an exception if any parameter value is invalid. Parameter value checks which do not depend on other parameters are handled by Param.validate().

Typical implementation should first conduct verification on schema change and parameter validity, including complex parameter interaction checks.

:: DeveloperApi ::

Derives the output schema from the input schema and parameters, optionally with logging.

This should be optimistic. If it is unclear whether the schema will be valid, then it should be assumed valid until proven otherwise.

An immutable unique ID for the object and its derivatives.

Validates and transforms the input schema with the provided param map.

Returns an MLWriter instance for this ML instance.