The CoFilter_2D object provides the convolution and pooling operators.
Attributes
- See also
-
mathstat.MatrixDfor infix implementations of conv (+) --validconvolution, no reversal (conv_ for reversal) convs (~+) --sameconvolution, with reversal convf (*++) --fullconvolution, with reversal - Companion
- class
- Graph
-
- Supertypes
-
class Objecttrait Matchableclass Any
- Self type
-
CoFilter_2D.type
Members list
Value members
Concrete methods
Return the 'valid' (no padding) convolution of cofilter matrix c and input matrix x. Caveat: does not include reversal.
Return the 'valid' (no padding) convolution of cofilter matrix c and input matrix x. Caveat: does not include reversal.
Value parameters
- c
-
the cofilter matrix of coefficients
- x
-
the input/data matrix
Attributes
Return the 'valid' (no padding) convolution of cofilter matrix c and input tensor x. Caveat: does not include reversal.
Return the 'valid' (no padding) convolution of cofilter matrix c and input tensor x. Caveat: does not include reversal.
Value parameters
- c
-
the cofilter matrix of coefficients
- x
-
the input/data tensor
Attributes
Return the 'full' convolution of cofilter matrix c and input matrix x.
Return the 'full' convolution of cofilter matrix c and input matrix x.
Value parameters
- c
-
the cofilter matrix of coefficients
- x
-
the input/data matrix
Attributes
Return the 'full' convolution of cofilter c matrix and input tensor x.
Return the 'full' convolution of cofilter c matrix and input tensor x.
Value parameters
- c
-
the cofilter matrix of coefficients
- x
-
the input/data tensor
Attributes
Return the 'same' (with padding) convolution of cofilter matrx c and input matrix x. Same means that the size of the result is the same as the input (via padding).
Return the 'same' (with padding) convolution of cofilter matrx c and input matrix x. Same means that the size of the result is the same as the input (via padding).
Value parameters
- c
-
the cofilter matrix of coefficients
- x
-
the input/data matrix
Attributes
Return the 'same' (with padding) convolution of cofilter matrix c and input tensor x. Same means that the size of the result is the same as the input (via padding).
Return the 'same' (with padding) convolution of cofilter matrix c and input tensor x. Same means that the size of the result is the same as the input (via padding).
Value parameters
- c
-
the cofilter matrix of coefficients
- x
-
the input/data tensor
Attributes
Return the max-pooling results over all pooling windows.
Return the max-pooling results over all pooling windows.
Value parameters
- s
-
the size (s x s) of the pooling window
- x
-
the input/data matrix
Attributes
Return the max-pooling results over all data instances (tensor level).
Return the max-pooling results over all data instances (tensor level).
Value parameters
- s
-
the the size of the pooling window
- x
-
the input/data tensor
Attributes
Return the avg-pooling results over all pooling windows.
Return the avg-pooling results over all pooling windows.
Value parameters
- s
-
the size (s x s) of the pooling window
- x
-
the input/data matrix