org.apache.ignite.ml.trainers

Class AdaptableDatasetModel<I,O,IW,OW,M extends IgniteModel<IW,OW>>

java.lang.Object

org.apache.ignite.ml.trainers.AdaptableDatasetModel<I,O,IW,OW,M>

Type Parameters:

I - Type of input of this model.

O - Type of output of this model.

IW - Type of input of inner model.

OW - Type of output of inner model.

M - Type of inner model.

All Implemented Interfaces:

Serializable, AutoCloseable, IgniteModel<I,O>, Model<I,O>

public class AdaptableDatasetModel<I,O,IW,OW,M extends IgniteModel<IW,OW>>
extends Object
implements IgniteModel<I,O>

Model which is composition of form before `andThen` inner Mdl `andThen` after.

See Also:

Serialized Form

Constructor Summary

Constructors

Constructor and Description
AdaptableDatasetModel(IgniteFunction<I,IW> before, M mdl, IgniteFunction<OW,O> after) Construct instance of this class.

Method Summary

Modifier and Type	Method and Description
<I1> AdaptableDatasetModel<I1,O,IW,OW,M>	andBefore(IgniteFunction<I1,I> before) Create new AdaptableDatasetModel which is a composition of the form thisMdl . before.
<O1> AdaptableDatasetModel<I,O1,IW,OW,M>	andThen(IgniteModel<O,O1> after) Get a composition model of the form x -> after(mdl(x)).
M	innerModel() Get inner model.
O	predict(I i) Result of this model application is a result of composition before `andThen` inner mdl `andThen` after.
<M1 extends IgniteModel<IW,OW>> AdaptableDatasetModel<I,O,IW,OW,M1>	withInnerModel(M1 mdl) Create new instance of this class with changed inner model.

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apache.ignite.ml.IgniteModel

andThen, close, combine, toString

Constructor Detail

AdaptableDatasetModel

public AdaptableDatasetModel(IgniteFunction<I,IW> before,
                             M mdl,
                             IgniteFunction<OW,O> after)

Construct instance of this class.

Parameters:

before - Function applied before wrapped model.

mdl - Inner model.

after - Function applied after wrapped model.

Method Detail

predict

public O predict(I i)

Result of this model application is a result of composition before `andThen` inner mdl `andThen` after.

Specified by:

predict in interface Model<I,O>

Parameters:

i - Input arguments.

Returns:

Prediction result.

andThen

public <O1> AdaptableDatasetModel<I,O1,IW,OW,M> andThen(IgniteModel<O,O1> after)

Get a composition model of the form x -> after(mdl(x)).

Specified by:

andThen in interface IgniteModel<I,O>

Type Parameters:

O1 - Type of output of function applied after this model.

Parameters:

after - Model to apply after this model.

Returns:

Composition model of the form x -> after(mdl(x)).

andBefore

public <I1> AdaptableDatasetModel<I1,O,IW,OW,M> andBefore(IgniteFunction<I1,I> before)

Create new AdaptableDatasetModel which is a composition of the form thisMdl . before.

Specified by:

andBefore in interface IgniteModel<I,O>

Type Parameters:

I1 - Type of function applied before this model.

Parameters:

before - Function applied before this model.

Returns:

New AdaptableDatasetModel which is a composition of the form thisMdl . before.

innerModel

public M innerModel()

Get inner model.

Returns:

Inner model.

withInnerModel

public <M1 extends IgniteModel<IW,OW>> AdaptableDatasetModel<I,O,IW,OW,M1> withInnerModel(M1 mdl)

Create new instance of this class with changed inner model.

Type Parameters:

M1 - Type of inner model.

Parameters:

mdl - Inner model.

Returns:

New instance of this class with changed inner model.