public abstract class BaseTransformFloatOp extends BaseTransformOp implements TransformFloatOp
dimensionz, extraArgz, x, xVertexId, y, yVertexId, z, zVertexIddimensions, extraArgs, inPlace, ownName, ownNameSetWithDefault, sameDiff, scalarValue| Constructor and Description |
|---|
BaseTransformFloatOp() |
BaseTransformFloatOp(INDArray x) |
BaseTransformFloatOp(INDArray x,
INDArray z) |
BaseTransformFloatOp(SameDiff sameDiff,
SDVariable i_v,
boolean inPlace) |
BaseTransformFloatOp(SameDiff sameDiff,
SDVariable i_v,
long[] shape,
boolean inPlace,
Object[] extraArgs) |
BaseTransformFloatOp(SameDiff sameDiff,
SDVariable i_v,
Object[] extraArgs) |
| Modifier and Type | Method and Description |
|---|---|
List<DataType> |
calculateOutputDataTypes(List<DataType> dataTypes)
Calculate the data types for the output arrays.
|
List<LongShapeDescriptor> |
calculateOutputShape()
Calculate the output shape for this op
|
List<LongShapeDescriptor> |
calculateOutputShape(OpContext oc) |
Op.Type |
getOpType() |
Op.Type |
opType()
The type of the op
|
DataType |
resultType()
This method returns datatype for result array wrt given inputs
|
DataType |
resultType(OpContext oc) |
boolean |
validateDataTypes(OpContext oc,
boolean experimentalMode) |
zclearArrays, defineDimensions, dimensions, equals, extraArgs, extraArgsBuff, extraArgsDataBuff, getFinalResult, getInputArgument, getNumOutputs, getOpType, hashCode, initFromOnnx, initFromTensorFlow, onnxName, outputVariables, setX, setY, setZ, tensorflowName, toCustomOp, toString, x, yarg, arg, argNames, args, attributeAdaptersForFunction, configFieldName, diff, doDiff, dup, getValue, isConfigProperties, larg, mappingsForFunction, onnxNames, opName, opNum, outputs, outputVariable, outputVariables, outputVariablesNames, propertiesForFunction, rarg, replaceArg, setInstanceId, setPropertiesForFunction, setValueFor, tensorflowNamesclone, finalize, getClass, notify, notifyAll, wait, wait, waitclearArrays, extraArgs, extraArgsBuff, extraArgsDataBuff, opName, opNum, setExtraArgs, setX, setY, setZ, toCustomOp, x, y, zpublic BaseTransformFloatOp(SameDiff sameDiff, SDVariable i_v, boolean inPlace)
public BaseTransformFloatOp(SameDiff sameDiff, SDVariable i_v, long[] shape, boolean inPlace, Object[] extraArgs)
public BaseTransformFloatOp(SameDiff sameDiff, SDVariable i_v, Object[] extraArgs)
public BaseTransformFloatOp()
public BaseTransformFloatOp(INDArray x)
public Op.Type getOpType()
getOpType in interface TransformOppublic Op.Type opType()
DifferentialFunctionopType in class DifferentialFunctionpublic DataType resultType()
TransformOpresultType in interface TransformOppublic DataType resultType(OpContext oc)
resultType in interface TransformOppublic boolean validateDataTypes(OpContext oc, boolean experimentalMode)
validateDataTypes in interface TransformOppublic List<LongShapeDescriptor> calculateOutputShape()
DifferentialFunctioncalculateOutputShape in class BaseTransformOppublic List<LongShapeDescriptor> calculateOutputShape(OpContext oc)
calculateOutputShape in class DifferentialFunctionpublic List<DataType> calculateOutputDataTypes(List<DataType> dataTypes)
DifferentialFunctionDifferentialFunction.calculateOutputShape(), this method differs in that it does not
require the input arrays to be populated.
This is important as it allows us to do greedy datatype inference for the entire net - even if arrays are not
available.calculateOutputDataTypes in class DifferentialFunctiondataTypes - The data types of the inputsCopyright © 2021. All rights reserved.