Python* API Reference for Intel® Data Analytics Acceleration Library 2020 Update 1

Public Member Functions | Static Public Attributes | List of all members
Parameter Class Reference

Parameters for the forward and backward pooling layers. More...

Public Member Functions

def __init__
 
- Public Member Functions inherited from Parameter
def __init__
 
- Public Member Functions inherited from Parameter
def __init__
 
def check
 

Static Public Attributes

 strides = ...
 
 paddings = ...
 
 kernelSizes = ...
 
 indices = ...
 
- Static Public Attributes inherited from Parameter
 weightsInitializer = ...
 
 biasesInitializer = ...
 
 predictionStage = ...
 
 propagateGradient = ...
 
 weightsAndBiasesInitialized = ...
 
 allowInplaceComputation = ...
 

Detailed Description

Deprecated:
This item will be removed in a future release.

Constructor & Destructor Documentation

def __init__ (   self,
  firstIndex,
  secondIndex,
  thirdIndex,
  firstKernelSize = 2,
  secondKernelSize = 2,
  thirdKernelSize = 2,
  firstStride = 2,
  secondStride = 2,
  thirdStride = 2,
  firstPadding = 0,
  secondPadding = 0,
  thirdPadding = 0 
)

Constructs the parameters of 3D pooling layer

Parameters
firstIndexIndex of the first of three dimensions on which the pooling is performed
secondIndexIndex of the second of three dimensions on which the pooling is performed
thirdIndexIndex of the third of three dimensions on which the pooling is performed
firstKernelSizeSize of the first dimension of three-dimensional subtensor for which the kernel is applied
secondKernelSizeSize of the second dimension of three-dimensional subtensor for which the kernel is applied
thirdKernelSizeSize of the third dimension of three-dimensional subtensor for which the kernel is applied
firstStrideInterval over the first dimension on which the pooling is performed
secondStrideInterval over the second dimension on which the pooling is performed
thirdStrideInterval over the third dimension on which the pooling is performed
firstPaddingNumber of data elements to implicitly add to the the first dimension of the three-dimensional subtensor on which the pooling is performed
secondPaddingNumber of data elements to implicitly add to the the second dimension of the three-dimensional subtensor on which the pooling is performed
thirdPaddingNumber of data elements to implicitly add to the the third dimension of the three-dimensional subtensor on which the pooling is performed
Deprecated:
This item will be removed in a future release.

Member Data Documentation

indices = ...
static

Indices of the three dimensions on which pooling is performed

kernelSizes = ...
static

Data structure representing the size of the three-dimensional subtensor

paddings = ...
static

Data structure representing the number of data elements to implicitly add

strides = ...
static

Data structure representing the intervals on which the subtensors for pooling are selected


The documentation for this class was generated from the following file:

For more complete information about compiler optimizations, see our Optimization Notice.