Modifier and Type | Method and Description |
---|---|
TransformProcess.Builder |
TransformProcess.Builder.transform(Transform transform)
Add a transformation to be executed after the previously-added operations have been executed
|
Constructor and Description |
---|
DataAction(Transform transform) |
Modifier and Type | Class and Description |
---|---|
class |
NDArrayColumnsMathOpTransform
Perform an element wise mathematical operation on 2 or more NDArray columns
|
class |
NDArrayDistanceTransform
Calculate the distance (cosine similarity, EUCLIDEAN, MANHATTAN) between two INDArrays
|
class |
NDArrayMathFunctionTransform
A simple transform to do common mathematical operations, such as sin(x), ceil(x), etc.
Operations are performed element-wise on each value in the INDArray; operations are specified by MathFunction |
class |
NDArrayScalarOpTransform
Perform an NDArray/scalar element wise operation, such as X.addi(scalar).
|
Modifier and Type | Class and Description |
---|---|
class |
ReduceSequenceTransform
Reduce The values in each column in the sequence to a single value using a reducer.
|
Modifier and Type | Class and Description |
---|---|
class |
BaseSequenceExpansionTransform
A base class for sequence expansion operations.
|
Modifier and Type | Class and Description |
---|---|
class |
SequenceTrimTransform
SequenceTrimTranform removes the first or last N values in a sequence.
|
Modifier and Type | Class and Description |
---|---|
class |
ReduceSequenceByWindowTransform
Idea: do two things.
|
Modifier and Type | Method and Description |
---|---|
Transform |
BaseSerializer.deserializeTransform(String str)
Deserialize a Transform serialized using
BaseSerializer.serialize(Object) |
Modifier and Type | Method and Description |
---|---|
List<Transform> |
BaseSerializer.deserializeTransformList(String str)
Deserialize a Transform List serialized using
BaseSerializer.serializeTransformList(List) , or
an array serialized using BaseSerializer.serialize(Transform[]) |
Modifier and Type | Method and Description |
---|---|
String |
BaseSerializer.serialize(Transform[] transforms) |
Modifier and Type | Method and Description |
---|---|
String |
BaseSerializer.serializeTransformList(List<Transform> list)
Serialize a list of Transforms
|
Constructor and Description |
---|
TransformList(List<Transform> list) |
Modifier and Type | Class and Description |
---|---|
class |
BaseColumnsMathOpTransform
Base class for multiple column math operations.
|
class |
BaseColumnTransform
Map the values in a single column to new values.
|
class |
BaseTransform
BaseTransform: an
abstract transform class, that handles transforming
sequences by transforming
each example individually
|
Modifier and Type | Class and Description |
---|---|
class |
CategoricalToIntegerTransform
Created by Alex on 4/03/2016.
|
class |
CategoricalToOneHotTransform
Created by Alex on 4/03/2016.
|
class |
IntegerToCategoricalTransform
Convert an integer column to a categorical column, using a provided
Map<Integer,String> |
class |
PivotTransform
Pivot transform operates on two columns:
- a categorical column that operates as a key, and
- Another column that contains a value
Essentially, Pivot transform takes keyvalue pairs and breaks them out into separate columns.
|
class |
StringToCategoricalTransform
Convert a String column
to a categorical column
|
Modifier and Type | Class and Description |
---|---|
class |
AddConstantColumnTransform
Add a new column, where the values in that column for all records are identical (according to the specified value)
|
class |
DuplicateColumnsTransform
Duplicate one or more columns.
|
class |
RemoveAllColumnsExceptForTransform
Transform that removes all columns except
for those that are explicitly
specified as ones to keep
To specify only the columns
to remove, use
RemoveColumnsTransform |
class |
RemoveColumnsTransform
Remove the specified columns from the data.
|
class |
RenameColumnsTransform
Rename one or more columns
|
class |
ReorderColumnsTransform
Rearrange the order of the columns.
|
Modifier and Type | Class and Description |
---|---|
class |
ConditionalCopyValueTransform
Replace the value in a specified column with a new value taken from another column, if a condition is satisfied/true.
Note that the condition can be any generic condition, including on other column(s), different to the column that will be modified if the condition is satisfied/true. |
class |
ConditionalReplaceValueTransform
Replace the value in a specified column with a new value, if a condition is satisfied/true.
Note that the condition can be any generic condition, including on other column(s), different to the column that will be modified if the condition is satisfied/true. |
class |
ConditionalReplaceValueTransformWithDefault
Replace the value in a specified column with a 'yes' value, if a condition is satisfied/true.
Replace the value of this same column with a 'no' value otherwise. |
Modifier and Type | Class and Description |
---|---|
class |
BaseDoubleTransform |
class |
ConvertToDouble
Convert any value to an Double
|
class |
DoubleColumnsMathOpTransform
Add a new double column, calculated from one or more other columns.
|
class |
DoubleMathFunctionTransform
A simple transform to do common mathematical operations, such as sin(x), ceil(x), etc.
Operations are specified by MathFunction |
class |
DoubleMathOpTransform
Double mathematical operation.
This is an in-place operation of the double column value and a double scalar. |
class |
Log2Normalizer
Normalize by taking scale * log2((in-columnMin)/(mean-columnMin) + 1)
Maps values in range (columnMin to infinity) to (0 to infinity)
Most suitable for values with a geometric/negative exponential type distribution.
|
class |
MinMaxNormalizer
Normalizer to map (min to max) -> (newMin-to newMax) linearly.
|
class |
StandardizeNormalizer
Normalize using (x-mean)/stdev.
|
class |
SubtractMeanNormalizer
Normalize by substracting the mean
|
Modifier and Type | Class and Description |
---|---|
class |
BaseIntegerTransform
Abstract integer transformation (single column)
|
class |
ConvertToInteger
Convert any value to an Integer.
|
class |
IntegerColumnsMathOpTransform
Add a new integer column, calculated from one or more other columns.
|
class |
IntegerMathOpTransform
Integer mathematical operation.
This is an in-place operation of the integer column value and an integer scalar. |
class |
IntegerToOneHotTransform
Convert an integer column to a set of one-hot columns.
|
class |
ReplaceEmptyIntegerWithValueTransform
Replace an empty/missing integer with a certain value.
|
class |
ReplaceInvalidWithIntegerTransform
Replace an invalid (non-integer) value in a column with a specified integer
|
Modifier and Type | Class and Description |
---|---|
class |
LongColumnsMathOpTransform
Add a new long column, calculated from one or more other columns.
|
class |
LongMathOpTransform
Long mathematical operation.
This is an in-place operation of the long column value and an long scalar. |
Modifier and Type | Class and Description |
---|---|
class |
TextToCharacterIndexTransform
Convert each text value in a sequence to a longer sequence of integer indices.
|
class |
TextToTermIndexSequenceTransform
Convert each text value in a sequence to a longer sequence of integer indices.
|
Modifier and Type | Class and Description |
---|---|
class |
ParseDoubleTransform
Convert string writables to doubles
|
Modifier and Type | Class and Description |
---|---|
class |
SequenceDifferenceTransform
SequenceDifferenceTransform: for an input sequence, calculate the difference on one column.
For each time t, calculate someColumn(t) - someColumn(t-s), where s >= 1 is the 'lookback' period. Note: at t=0 (i.e., the first step in a sequence; or more generally, for all times t < s), there is no previous value from which to calculate the difference. |
class |
SequenceMovingWindowReduceTransform
SequenceMovingWindowReduceTransform Adds a new column, where the value is derived by:
(a) using a window of the last N values in a single column, (b) Apply a reduction op on the window to calculate a new value for example, this transformer can be used to implement a simple moving average of the last N values, or determine the minimum or maximum values in the last N time steps. |
class |
SequenceOffsetTransform
Sequence offset transform takes a sequence, and shifts The values in one or more columns by a specified number of
times steps.
|
Modifier and Type | Class and Description |
---|---|
class |
AppendStringColumnTransform
Append a String to the
values in a single column
|
class |
BaseStringTransform
Abstract String column transform
|
class |
ChangeCaseStringTransform
Change case (to, e.g, all lower case) of String column.
|
class |
ConcatenateStringColumns
Concatenate values of one or more String columns into
a new String column.
|
class |
ConvertToString
Convert any value to a string.
|
class |
MapAllStringsExceptListTransform
This method maps all String values, except those is the specified list, to a single String value
|
class |
RemoveWhiteSpaceTransform
String transform that removes all whitespace charaters
|
class |
ReplaceEmptyStringTransform
Replace empty String values with the specified String
|
class |
ReplaceStringTransform
Replaces String values that match regular expressions.
|
class |
StringListToCategoricalSetTransform
Convert a delimited String to a list of binary categorical columns.
|
class |
StringListToCountsNDArrayTransform
Converts String column into a bag-of-words (BOW) represented as an NDArray of "counts."
Note that the original column is removed in the process |
class |
StringListToIndicesNDArrayTransform
Converts String column into a sparse bag-of-words (BOW)
represented as an NDArray of indices.
|
class |
StringMapTransform
A simple String -> String map function.
|
Modifier and Type | Class and Description |
---|---|
class |
DeriveColumnsFromTimeTransform
Create a number of new columns by deriving their values from a Time column.
|
class |
StringToTimeTransform
Convert a String column to a time column by parsing the date/time String, using a JodaTime.
|
class |
TimeMathOpTransform
Transform math op on a time column
|
Copyright © 2018. All rights reserved.