ebl::maxss_module< T, Tstate > Class Template Reference
This module applies max subsampling. More...
#include <ebl_pooling.h>
Inheritance diagram for ebl::maxss_module< T, Tstate >:
Public Member Functions | |
| maxss_module (uint thickness, idxdim &kernel, idxdim &stride, const char *name="maxss") | |
| virtual | ~maxss_module () |
| Destructor. | |
| virtual void | fprop (Tstate &in, Tstate &out) |
| forward propagation from in to out | |
| virtual void | bprop (Tstate &in, Tstate &out) |
| backward propagation from out to in | |
| virtual void | bbprop (Tstate &in, Tstate &out) |
| second-derivative backward propagation from out to in | |
| virtual bool | resize_output (Tstate &in, Tstate &out) |
| virtual int | replicable_order () |
| order of operation | |
| virtual fidxdim | fprop_size (fidxdim &i_size) |
| virtual fidxdim | bprop_size (const fidxdim &o_size) |
| virtual maxss_module< T, Tstate > * | copy (parameter< T, Tstate > *p=NULL) |
| virtual std::string | describe () |
| Returns a string describing this module and its parameters. | |
Protected Attributes | |
| uint | thickness |
| Number of features. | |
| idxdim | kernel |
| Kernel dimensions (1st dim is thickness). | |
| idxdim | stride |
| Stride dimensions (stride 1 in 1st dim). | |
| idx< int > | switches |
| Remember max locations. | |
| bool | float_precision |
| check the precision of the module | |
| bool | double_precision |
| check the precision of the module | |
| idx< T > | indices |
| Remember max locations. | |
Detailed Description
template<typename T, class Tstate = bbstate_idx<T>>
class ebl::maxss_module< T, Tstate >
This module applies max subsampling.
Constructor & Destructor Documentation
template<typename T , class Tstate >
| ebl::maxss_module< T, Tstate >::maxss_module | ( | uint | thickness, |
| idxdim & | kernel, | ||
| idxdim & | stride, | ||
| const char * | name = "maxss" |
||
| ) |
Constructor.
- Parameters:
-
thickness The number of features. kernel Size of subsampling kernel (without thickness). stride Stride of subsampling kernel (without thickness).
Member Function Documentation
template<typename T , class Tstate >
| fidxdim ebl::maxss_module< T, Tstate >::bprop_size | ( | const fidxdim & | o_size | ) | [virtual] |
Return dimensions compatible with this module given output dimensions. See module_1_1_gen's documentation for more details.
Just multiply each dimension by its stride
Reimplemented from ebl::module_1_1< T, Tstate >.
template<typename T , class Tstate >
| maxss_module< T, Tstate > * ebl::maxss_module< T, Tstate >::copy | ( | parameter< T, Tstate > * | p = NULL | ) | [virtual] |
Returns a deep copy of this module.
- Parameters:
-
p If NULL, reuse current parameter space, otherwise allocate new weights on parameter 'p'.
Reimplemented from ebl::module_1_1< T, Tstate >.
template<typename T , class Tstate >
| fidxdim ebl::maxss_module< T, Tstate >::fprop_size | ( | fidxdim & | i_size | ) | [virtual] |
Return dimensions that are compatible with this module. See module_1_1_gen's documentation for more details.
Recompute the input size to be compliant with the output
Reimplemented from ebl::module_1_1< T, Tstate >.
template<typename T , class Tstate >
| bool ebl::maxss_module< T, Tstate >::resize_output | ( | Tstate & | in, |
| Tstate & | out | ||
| ) | [virtual] |
resize the output based on input dimensions This returns true if output was resized/reallocated, false otherwise.
The documentation for this class was generated from the following files:
- /home/rex/ebltrunk/core/libeblearn/include/ebl_pooling.h
- /home/rex/ebltrunk/core/libeblearn/include/ebl_pooling.hpp
