Class that specifies parameters of the 2D locally connected layer.
More...
long getGroupDimension |
( |
| ) |
|
Gets the index of the dimension for which the grouping is applied
- Returns
- LocallyConnected2dIndex of the dimension for which the grouping is applied
Gets the data structure representing the dimension for locally connected kernels
- Returns
- Data structure representing the dimension for locally connected kernels
Gets the data structure representing the sizes of the two-dimensional kernel subtensor
- Returns
- Data structure representing the sizes of the two-dimensional kernel subtensor
Gets the number of groups which the input data is split in groupDimension dimension
- Returns
- Number of groups which the input data is split in groupDimension dimension
Gets the number of kernels applied to the input layer data
- Returns
- Number of kernels applied to the input layer data
Gets the data structure representing the number of data elements to implicitly add to each side of the 2D subtensor on which locally connected is performed
- Returns
- Data structure representing the number of data elements to implicitly add to each side of the 2D subtensor on which locally connected is performed
Gets the data structure representing the intervals on which the kernel should be applied to the input
- Returns
- Data structure representing the intervals on which the kernel should be applied to the input
void setGroupDimension |
( |
long |
groupDimension | ) |
|
Sets the index of the dimension for which the grouping is applied
- Parameters
-
groupDimension | Dimension for which the grouping is applied |
Sets the data structure representing the dimension for locally connected kernels
- Parameters
-
Sets the data structure representing the sizes of the two-dimensional kernel subtensor
- Parameters
-
kernelSize | The data structure representing the sizes of the two-dimensional kernel subtensor |
void setNGroups |
( |
long |
nGroups | ) |
|
Sets the number of groups which the input data is split in groupDimension dimension
- Parameters
-
nGroups | The number of groups which the input data is split in groupDimension dimension |
void setNKernels |
( |
long |
nKernels | ) |
|
Sets the number of kernels applied to the input layer data
- Parameters
-
nKernels | The number of kernels applied to the input layer data |
Sets the data structure representing the number of data elements to implicitly add to each side of the 2D subtensor on which locally connected is performed
- Parameters
-
padding | The data structure representing the number of data elements to implicitly add to each side of the 2D subtensor on which locally connected is performed |
Sets the data structure representing the intervals on which the kernel should be applied to the input
- Parameters
-
str | The data structure representing the intervals on which the kernel should be applied to the input |
The documentation for this class was generated from the following file:
- LocallyConnected2dParameter.java