org.tribuo.hash

Class HashingTrainer<T extends Output<T>>

java.lang.Object

org.tribuo.hash.HashingTrainer<T>

Type Parameters:

T - The type of Output this trainer works with.

All Implemented Interfaces:

com.oracle.labs.mlrg.olcut.config.Configurable, com.oracle.labs.mlrg.olcut.provenance.Provenancable<TrainerProvenance>, Trainer<T>

public final class HashingTrainer<T extends Output<T>>
extends Object
implements Trainer<T>

A Trainer which hashes the Dataset before the Model is produced. This means the model does not contain any feature names, only one way hashes of names.

It wraps another Trainer which actually performs the training.

Field Summary

Fields inherited from interface org.tribuo.Trainer

DEFAULT_SEED

Constructor Summary

Constructors

Constructor and Description
HashingTrainer(Trainer<T> trainer, Hasher hasher)

Method Summary

Modifier and Type	Method and Description
int	getInvocationCount() The number of times this trainer instance has had it's train method invoked.
TrainerProvenance	getProvenance()
Model<T>	train(Dataset<T> dataset, Map<String,com.oracle.labs.mlrg.olcut.provenance.Provenance> instanceProvenance) This clones the Dataset, hashes each of the examples and rewrites their feature ids before passing it to the inner trainer.

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.tribuo.Trainer

train

Methods inherited from interface com.oracle.labs.mlrg.olcut.config.Configurable

postConfig

Constructor Detail

HashingTrainer

public HashingTrainer(Trainer<T> trainer,
                      Hasher hasher)

Method Detail

train

public Model<T> train(Dataset<T> dataset,
                      Map<String,com.oracle.labs.mlrg.olcut.provenance.Provenance> instanceProvenance)

This clones the Dataset, hashes each of the examples and rewrites their feature ids before passing it to the inner trainer.

This ensures the Trainer sees the data after the collisions, and thus builds the correct size data structures.

Specified by:

train in interface Trainer<T extends Output<T>>

Parameters:

dataset - The input dataset.

instanceProvenance - Provenance information specific to this execution of train (e.g., cross validation fold number).

Returns:

A trained Model.

getInvocationCount

public int getInvocationCount()

Description copied from interface: Trainer

The number of times this trainer instance has had it's train method invoked.

This is used to determine how many times the trainer's RNG has been accessed to ensure replicability in the random number stream.

Specified by:

getInvocationCount in interface Trainer<T extends Output<T>>

Returns:

The number of train invocations.

getProvenance

public TrainerProvenance getProvenance()

Specified by:

getProvenance in interface com.oracle.labs.mlrg.olcut.provenance.Provenancable<TrainerProvenance>