water.api.schemas3

Class ModelParametersSchemaV3<P extends Model.Parameters,S extends ModelParametersSchemaV3<P,S>>

java.lang.Object

water.Iced

water.api.Schema<I,S>

water.api.schemas3.SchemaV3<P,S>

water.api.schemas3.ModelParametersSchemaV3<P,S>

All Implemented Interfaces:

java.io.Externalizable, java.io.Serializable, java.lang.Cloneable, Freezable

Direct Known Subclasses:

ClusteringModelParametersSchemaV3, QuantileV3.QuantileParametersV3

public class ModelParametersSchemaV3<P extends Model.Parameters,S extends ModelParametersSchemaV3<P,S>>
extends SchemaV3<P,S>

An instance of a ModelParameters schema contains the Model build parameters (e.g., K and max_iterations for KMeans). NOTE: use subclasses, not this class directly. It is not abstract only so that we can instantiate it to generate metadata for it for the metadata API.

See Also:

Serialized Form

Nested Class Summary

Nested classes/interfaces inherited from class water.api.schemas3.SchemaV3

SchemaV3.Meta

Nested classes/interfaces inherited from class water.api.Schema

Schema.AutoParseable

Field Summary

Fields

Modifier and Type	Field and Description
MultinomialAucType	auc_type
Model.Parameters.CategoricalEncodingScheme	categorical_encoding
KeyV3.ModelKeyV3	checkpoint A model key associated with a previously trained model.
java.lang.String	custom_distribution_func
java.lang.String	custom_metric_func
hex.genmodel.utils.DistributionFamily	distribution
java.lang.String	export_checkpoints_dir
Model.Parameters.FoldAssignmentScheme	fold_assignment
FrameV3.ColSpecifierV3	fold_column
int	gainslift_bins
double	huber_alpha
boolean	ignore_const_cols
java.lang.String[]	ignored_columns
boolean	keep_cross_validation_fold_assignment
boolean	keep_cross_validation_models
boolean	keep_cross_validation_predictions
int	max_categorical_levels
double	max_runtime_secs
KeyV3.ModelKeyV3	model_id
int	nfolds
FrameV3.ColSpecifierV3	offset_column
boolean	parallelize_cross_validation
double	quantile_alpha
FrameV3.ColSpecifierV3	response_column
boolean	score_each_iteration
ScoreKeeper.StoppingMetric	stopping_metric Metric to use for convergence checking, only for _stopping_rounds > 0
int	stopping_rounds Early stopping based on convergence of stopping_metric.
double	stopping_tolerance
KeyV3.FrameKeyV3	training_frame
double	tweedie_power
KeyV3.FrameKeyV3	validation_frame
FrameV3.ColSpecifierV3	weights_column

Fields inherited from class water.api.schemas3.SchemaV3

__meta

Constructor Summary

Constructors

Constructor and Description
ModelParametersSchemaV3()

Method Summary

Modifier and Type	Method and Description
protected static java.lang.String[]	append_field_arrays(java.lang.String[] first, java.lang.String[] second)
protected static <X extends java.lang.Class<? extends ModelParametersSchemaV3>> java.util.List<java.lang.String>	extractDeclaredApiParameters(X schemaClass)
java.lang.String[]	fields()
S	fillFromImpl(P impl) Fill this Schema from the given implementation object.
P	fillImpl(P impl) Fill an impl object and any children from this schema and its children.
protected ModelParameterSchemaV3[]	getAdditionalParameters()
static AutoBuffer	writeParametersJSON(AutoBuffer ab, ModelParametersSchemaV3 parameters, ModelParametersSchemaV3 input_parameters, ModelParametersSchemaV3 default_parameters, java.lang.String name) Write the parameters, including their metadata, into an AutoBuffer.

Methods inherited from class water.api.schemas3.SchemaV3

writeJSON

Methods inherited from class water.api.Schema

createAndFillImpl, createImpl, extractVersionFromSchemaName, fillFromAny, fillFromBody, fillFromImpl, fillFromImpl, fillFromParms, fillFromParms, fillFromParms, fillImpl, getImplClass, getImplClass, getSchemaName, getSchemaType, getSchemaVersion, init_meta, markdown, markdown, newInstance, newInstance, setField, setSchemaType_doNotCall

Methods inherited from class water.Iced

asBytes, clone, copyOver, frozenType, read, readExternal, readJSON, reloadFromBytes, toJsonBytes, toJsonString, write, writeExternal

Methods inherited from class java.lang.Object

equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Field Detail

model_id

@API(level=critical,
     direction=INOUT,
     help="Destination id for this model; auto-generated if not specified.")
public KeyV3.ModelKeyV3 model_id

training_frame

@API(level=critical,
     direction=INOUT,
     help="Id of the training data frame.")
public KeyV3.FrameKeyV3 training_frame

validation_frame

@API(level=critical,
     direction=INOUT,
     gridable=true,
     help="Id of the validation data frame.")
public KeyV3.FrameKeyV3 validation_frame

nfolds

@API(level=critical,
     direction=INOUT,
     help="Number of folds for K-fold cross-validation (0 to disable or >= 2).")
public int nfolds

keep_cross_validation_models

@API(level=expert,
     direction=INOUT,
     help="Whether to keep the cross-validation models.")
public boolean keep_cross_validation_models

keep_cross_validation_predictions

@API(level=expert,
     direction=INOUT,
     help="Whether to keep the predictions of the cross-validation models.")
public boolean keep_cross_validation_predictions

keep_cross_validation_fold_assignment

@API(level=expert,
     direction=INOUT,
     help="Whether to keep the cross-validation fold assignment.")
public boolean keep_cross_validation_fold_assignment

parallelize_cross_validation

@API(help="Allow parallel training of cross-validation models",
     direction=INOUT,
     level=expert)
public boolean parallelize_cross_validation

distribution

@API(help="Distribution function",
     values={"AUTO","bernoulli","quasibinomial","ordinal","multinomial","gaussian","poisson","gamma","tweedie","laplace","quantile","huber","custom"},
     level=secondary,
     gridable=true)
public hex.genmodel.utils.DistributionFamily distribution

tweedie_power

@API(level=secondary,
     direction=INPUT,
     gridable=true,
     help="Tweedie power for Tweedie regression, must be between 1 and 2.")
public double tweedie_power

quantile_alpha

@API(level=secondary,
     direction=INPUT,
     gridable=true,
     help="Desired quantile for Quantile regression, must be between 0 and 1.")
public double quantile_alpha

huber_alpha

@API(help="Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be between 0 and 1).",
     level=secondary,
     direction=INPUT,
     gridable=true)
public double huber_alpha

response_column

@API(level=critical,
     direction=INOUT,
     gridable=true,
     is_member_of_frames={"training_frame","validation_frame"},
     is_mutually_exclusive_with="ignored_columns",
     help="Response variable column.")
public FrameV3.ColSpecifierV3 response_column

weights_column

@API(level=secondary,
     direction=INOUT,
     gridable=true,
     is_member_of_frames={"training_frame","validation_frame"},
     is_mutually_exclusive_with={"ignored_columns","response_column"},
     help="Column with observation weights. Giving some observation a weight of zero is equivalent to excluding it from the dataset; giving an observation a relative weight of 2 is equivalent to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row observation weights and do not increase the size of the data frame. This is typically the number of times a row is repeated, but non-integer values are supported as well. During training, rows with higher weights matter more, due to the larger loss function pre-factor. If you set weight = 0 for a row, the returned prediction frame at that row is zero and this is incorrect. To get an accurate prediction, remove all rows with weight == 0.")
public FrameV3.ColSpecifierV3 weights_column

offset_column

@API(level=secondary,
     direction=INOUT,
     gridable=true,
     is_member_of_frames={"training_frame","validation_frame"},
     is_mutually_exclusive_with={"ignored_columns","response_column","weights_column"},
     help="Offset column. This will be added to the combination of columns before applying the link function.")
public FrameV3.ColSpecifierV3 offset_column

fold_column

@API(level=secondary,
     direction=INOUT,
     gridable=true,
     is_member_of_frames="training_frame",
     is_mutually_exclusive_with={"ignored_columns","response_column","weights_column","offset_column"},
     help="Column with cross-validation fold index assignment per observation.")
public FrameV3.ColSpecifierV3 fold_column

fold_assignment

@API(level=secondary,
     direction=INOUT,
     gridable=true,
     values={"AUTO","Random","Modulo","Stratified"},
     help="Cross-validation fold assignment scheme, if fold_column is not specified. The \'Stratified\' option will stratify the folds based on the response variable, for classification problems.")
public Model.Parameters.FoldAssignmentScheme fold_assignment

categorical_encoding

@API(level=secondary,
     direction=INOUT,
     gridable=true,
     values={"AUTO","Enum","OneHotInternal","OneHotExplicit","Binary","Eigen","LabelEncoder","SortByResponse","EnumLimited"},
     help="Encoding scheme for categorical features")
public Model.Parameters.CategoricalEncodingScheme categorical_encoding

max_categorical_levels

@API(level=secondary,
     direction=INPUT,
     gridable=true,
     help="For every categorical feature, only use this many most frequent categorical levels for model training. Only used for categorical_encoding == EnumLimited.")
public int max_categorical_levels

ignored_columns

@API(level=critical,
     direction=INOUT,
     is_member_of_frames={"training_frame","validation_frame"},
     help="Names of columns to ignore for training.")
public java.lang.String[] ignored_columns

ignore_const_cols

@API(level=critical,
     direction=INOUT,
     help="Ignore constant columns.")
public boolean ignore_const_cols

score_each_iteration

@API(level=secondary,
     direction=INOUT,
     help="Whether to score during each iteration of model training.")
public boolean score_each_iteration

checkpoint

@API(level=secondary,
     direction=INOUT,
     help="Model checkpoint to resume training with.")
public KeyV3.ModelKeyV3 checkpoint

A model key associated with a previously trained model. This option allows users to build a new model as a continuation of a previously generated model (e.g., by a grid search).

stopping_rounds

@API(help="Early stopping based on convergence of stopping_metric. Stop if simple moving average of length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0 to disable)",
     level=secondary,
     direction=INOUT,
     gridable=true)
public int stopping_rounds

Early stopping based on convergence of stopping_metric. Stop if simple moving average of length k of the stopping_metric does not improve (by stopping_tolerance) for k=stopping_rounds scoring events." Can only trigger after at least 2k scoring events. Use 0 to disable.

max_runtime_secs

@API(help="Maximum allowed runtime in seconds for model training. Use 0 to disable.",
     level=secondary,
     direction=INOUT,
     gridable=true)
public double max_runtime_secs

stopping_metric

@API(help="Metric to use for early stopping (AUTO: logloss for classification, deviance for regression and anomaly_score for Isolation Forest). Note that custom and custom_increasing can only be used in GBM and DRF with the Python client.",
     valuesProvider=ModelParamsValuesProviders.StoppingMetricValuesProvider.class,
     level=secondary,
     direction=INOUT,
     gridable=true)
public ScoreKeeper.StoppingMetric stopping_metric

Metric to use for convergence checking, only for _stopping_rounds > 0

stopping_tolerance

@API(help="Relative tolerance for metric-based stopping criterion (stop if relative improvement is not at least this much)",
     level=secondary,
     direction=INOUT,
     gridable=true)
public double stopping_tolerance

gainslift_bins

@API(help="Gains/Lift table number of bins. 0 means disabled.. Default value -1 means automatic binning.",
     level=secondary,
     direction=INOUT)
public int gainslift_bins

custom_metric_func

@API(help="Reference to custom evaluation function, format: `language:keyName=funcName`",
     level=secondary,
     direction=INOUT,
     gridable=false)
public java.lang.String custom_metric_func

custom_distribution_func

@API(help="Reference to custom distribution, format: `language:keyName=funcName`",
     level=secondary,
     direction=INOUT,
     gridable=false)
public java.lang.String custom_distribution_func

export_checkpoints_dir

@API(help="Automatically export generated models to this directory.",
     level=secondary,
     direction=INOUT)
public java.lang.String export_checkpoints_dir

auc_type

@API(help="Set default multinomial AUC type.",
     valuesProvider=ModelParamsValuesProviders.MultinomialAucTypeSchemeValuesProvider.class,
     level=secondary,
     direction=INOUT,
     gridable=true)
public MultinomialAucType auc_type

Constructor Detail

ModelParametersSchemaV3

public ModelParametersSchemaV3()

Method Detail

fields

public java.lang.String[] fields()

append_field_arrays

protected static java.lang.String[] append_field_arrays(java.lang.String[] first,
                                                        java.lang.String[] second)

fillFromImpl

public S fillFromImpl(P impl)

Description copied from class: Schema

Fill this Schema from the given implementation object. If a schema doesn't need to adapt any fields if does not need to override this method.

Overrides:

fillFromImpl in class Schema<P extends Model.Parameters,S extends ModelParametersSchemaV3<P,S>>

fillImpl

public P fillImpl(P impl)

Description copied from class: Schema

Fill an impl object and any children from this schema and its children. If a schema doesn't need to adapt any fields if does not need to override this method.

Overrides:

fillImpl in class Schema<P extends Model.Parameters,S extends ModelParametersSchemaV3<P,S>>

writeParametersJSON

public static AutoBuffer writeParametersJSON(AutoBuffer ab,
                                             ModelParametersSchemaV3 parameters,
                                             ModelParametersSchemaV3 input_parameters,
                                             ModelParametersSchemaV3 default_parameters,
                                             java.lang.String name)

Write the parameters, including their metadata, into an AutoBuffer. Used by ModelBuilderSchema#writeJSON_impl and ModelSchemaV3#writeJSON_impl.

getAdditionalParameters

protected ModelParameterSchemaV3[] getAdditionalParameters()

extractDeclaredApiParameters

protected static <X extends java.lang.Class<? extends ModelParametersSchemaV3>> java.util.List<java.lang.String> extractDeclaredApiParameters(X schemaClass)

Type Parameters:

X - A generic type for a Class object representing a class extending ModelParametersSchemaV3.

Parameters:

schemaClass - A schema class to extract API annotated parameters

Returns: