public class MinMaxStrategy extends Object implements NormalizerStrategy<MinMaxStats>, Serializable
NormalizerStrategy
implementation that will normalize and denormalize data arrays to a given range, based on
statistics of the upper and lower bounds of the populationConstructor and Description |
---|
MinMaxStrategy() |
MinMaxStrategy(double minRange,
double maxRange) |
Modifier and Type | Method and Description |
---|---|
NormalizerStats.Builder |
newStatsBuilder()
Create a new
NormalizerStats.Builder instance that can be used to fit new data and of the type that
belongs to the current NormalizerStrategy implementation |
void |
preProcess(INDArray array,
INDArray maskArray,
MinMaxStats stats)
Normalize a data array
|
void |
revert(INDArray array,
INDArray maskArray,
MinMaxStats stats)
Denormalize a data array
|
public MinMaxStrategy()
public MinMaxStrategy(double minRange, double maxRange)
minRange
- the target range lower boundmaxRange
- the target range upper boundpublic void preProcess(INDArray array, INDArray maskArray, MinMaxStats stats)
preProcess
in interface NormalizerStrategy<MinMaxStats>
array
- the data to normalizestats
- statistics of the data populationpublic void revert(INDArray array, INDArray maskArray, MinMaxStats stats)
revert
in interface NormalizerStrategy<MinMaxStats>
array
- the data to denormalizestats
- statistics of the data populationpublic NormalizerStats.Builder newStatsBuilder()
NormalizerStats.Builder
instance that can be used to fit new data and of the type that
belongs to the current NormalizerStrategy implementationnewStatsBuilder
in interface NormalizerStrategy<MinMaxStats>
Copyright © 2017. All rights reserved.