Pad
Pads tensor.
nnf_pad(input, pad, mode = "constant", value = 0)
input |
(Tensor) N-dimensional tensor |
pad |
(tuple) m-elements tuple, where \frac{m}{2} ≤q input dimensions and m is even. |
mode |
'constant', 'reflect', 'replicate' or 'circular'. Default: 'constant' |
value |
fill value for 'constant' padding. Default: 0. |
The padding size by which to pad some dimensions of input
are described starting from the last dimension and moving forward.
≤ft\lfloor\frac{\mbox{len(pad)}}{2}\right\rfloor dimensions
of input will be padded.
For example, to pad only the last dimension of the input tensor, then
pad has the form
(\mbox{padding\_left}, \mbox{padding\_right});
to pad the last 2 dimensions of the input tensor, then use
(\mbox{padding\_left}, \mbox{padding\_right},
\mbox{padding\_top}, \mbox{padding\_bottom});
to pad the last 3 dimensions, use
(\mbox{padding\_left}, \mbox{padding\_right},
\mbox{padding\_top}, \mbox{padding\_bottom}
\mbox{padding\_front}, \mbox{padding\_back}).
See nn_constant_pad_2d, nn_reflection_pad_2d, and
nn_replication_pad_2d for concrete examples on how each of the
padding modes works. Constant padding is implemented for arbitrary dimensions.
tensor, or the last 2 dimensions of 4D input tensor, or the last dimension of
3D input tensor. Reflect padding is only implemented for padding the last 2
dimensions of 4D input tensor, or the last dimension of 3D input tensor.
Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.