forked from TensorLayer/tensorlayer3
215 lines
7.1 KiB
Python
215 lines
7.1 KiB
Python
#! /usr/bin/python
|
|
# -*- coding: utf-8 -*-
|
|
|
|
import tensorlayer as tl
|
|
from tensorlayer import logging
|
|
from tensorlayer.layers.core import Module
|
|
|
|
__all__ = [
|
|
'SubpixelConv1d',
|
|
'SubpixelConv2d',
|
|
]
|
|
|
|
|
|
class SubpixelConv1d(Module):
|
|
"""It is a 1D sub-pixel up-sampling layer.
|
|
|
|
Calls a TensorFlow function that directly implements this functionality.
|
|
We assume input has dim (batch, width, r)
|
|
|
|
Parameters
|
|
------------
|
|
scale : int
|
|
The up-scaling ratio, a wrong setting will lead to Dimension size error.
|
|
act : activation function
|
|
The activation function of this layer.
|
|
in_channels : int
|
|
The number of in channels.
|
|
name : str
|
|
A unique layer name.
|
|
|
|
Examples
|
|
----------
|
|
With TensorLayer
|
|
|
|
>>> net = tl.layers.Input([8, 25, 32], name='input')
|
|
>>> subpixelconv1d = tl.layers.SubpixelConv1d(scale=2, name='subpixelconv1d')(net)
|
|
>>> print(subpixelconv1d)
|
|
>>> output shape : (8, 50, 16)
|
|
|
|
References
|
|
-----------
|
|
`Audio Super Resolution Implementation <https://github.com/kuleshov/audio-super-res/blob/master/src/models/layers/subpixel.py>`__.
|
|
|
|
"""
|
|
|
|
def __init__(
|
|
self,
|
|
scale=2,
|
|
act=None,
|
|
in_channels=None,
|
|
name=None # 'subpixel_conv1d'
|
|
):
|
|
super().__init__(name, act=act)
|
|
self.scale = scale
|
|
self.in_channels = in_channels
|
|
# self.out_channels = int(self.in_channels / self.scale)
|
|
|
|
if self.in_channels is not None:
|
|
self.build(None)
|
|
self._built = True
|
|
|
|
logging.info(
|
|
"SubpixelConv1d %s: scale: %d act: %s" %
|
|
(self.name, scale, self.act.__class__.__name__ if self.act is not None else 'No Activation')
|
|
)
|
|
|
|
def __repr__(self):
|
|
actstr = self.act.__class__.__name__ if self.act is not None else 'No Activation'
|
|
s = ('{classname}(in_channels={in_channels}, out_channels={out_channels}')
|
|
s += (', ' + actstr)
|
|
if self.name is not None:
|
|
s += ', name=\'{name}\''
|
|
s += ')'
|
|
return s.format(classname=self.__class__.__name__, **self.__dict__)
|
|
|
|
def build(self, inputs_shape):
|
|
if inputs_shape is not None:
|
|
self.in_channels = inputs_shape[-1]
|
|
self.out_channels = int(self.in_channels / self.scale)
|
|
self.transpose = tl.ops.Transpose(perm=[2, 1, 0])
|
|
self.batch_to_space = tl.ops.BatchToSpace(block_size=[self.scale], crops=[[0, 0]])
|
|
|
|
def forward(self, inputs):
|
|
if self._forward_state == False:
|
|
if self._built == False:
|
|
self.build(tl.get_tensor_shape(inputs))
|
|
self._built = True
|
|
self._forward_state = True
|
|
|
|
outputs = self._PS(inputs)
|
|
if self.act is not None:
|
|
outputs = self.act(outputs)
|
|
return outputs
|
|
|
|
def _PS(self, I):
|
|
X = self.transpose(I) # (r, w, b)
|
|
X = self.batch_to_space(X) # (1, r*w, b)
|
|
X = self.transpose(X)
|
|
return X
|
|
|
|
|
|
class SubpixelConv2d(Module):
|
|
"""It is a 2D sub-pixel up-sampling layer, usually be used
|
|
for Super-Resolution applications, see `SRGAN <https://github.com/tensorlayer/srgan/>`__ for example.
|
|
|
|
Parameters
|
|
------------
|
|
scale : int
|
|
The up-scaling ratio, a wrong setting will lead to dimension size error.
|
|
n_out_channel : int or None
|
|
The number of output channels.
|
|
- If None, automatically set n_out_channel == the number of input channels / (scale x scale).
|
|
- The number of input channels == (scale x scale) x The number of output channels.
|
|
act : activation function
|
|
The activation function of this layer.
|
|
in_channels : int
|
|
The number of in channels.
|
|
name : str
|
|
A unique layer name.
|
|
|
|
Examples
|
|
---------
|
|
With TensorLayer
|
|
|
|
>>> # examples here just want to tell you how to set the n_out_channel.
|
|
>>> net = tl.layers.Input([2, 16, 16, 4], name='input1')
|
|
>>> subpixelconv2d = tl.layers.SubpixelConv2d(scale=2, n_out_channels=1, name='subpixel_conv2d1')(net)
|
|
>>> print(subpixelconv2d)
|
|
>>> output shape : (2, 32, 32, 1)
|
|
|
|
>>> net = tl.layers.Input([2, 16, 16, 4*10], name='input2')
|
|
>>> subpixelconv2d = tl.layers.SubpixelConv2d(scale=2, n_out_channels=10, name='subpixel_conv2d2')(net)
|
|
>>> print(subpixelconv2d)
|
|
>>> output shape : (2, 32, 32, 10)
|
|
|
|
>>> net = tl.layers.Input([2, 16, 16, 25*10], name='input3')
|
|
>>> subpixelconv2d = tl.layers.SubpixelConv2d(scale=5, n_out_channels=10, name='subpixel_conv2d3')(net)
|
|
>>> print(subpixelconv2d)
|
|
>>> output shape : (2, 80, 80, 10)
|
|
|
|
References
|
|
------------
|
|
- `Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network <https://arxiv.org/pdf/1609.05158.pdf>`__
|
|
|
|
"""
|
|
|
|
# github/Tetrachrome/subpixel https://github.com/Tetrachrome/subpixel/blob/master/subpixel.py
|
|
def __init__(
|
|
self,
|
|
scale=2,
|
|
n_out_channels=None,
|
|
act=None,
|
|
in_channels=None,
|
|
name=None # 'subpixel_conv2d'
|
|
):
|
|
super().__init__(name, act=act)
|
|
self.scale = scale
|
|
self.n_out_channels = n_out_channels
|
|
self.in_channels = in_channels
|
|
|
|
if self.in_channels is not None:
|
|
self.build(None)
|
|
self._built = True
|
|
logging.info(
|
|
"SubpixelConv2d %s: scale: %d act: %s" %
|
|
(self.name, scale, self.act.__class__.__name__ if self.act is not None else 'No Activation')
|
|
)
|
|
|
|
def __repr__(self):
|
|
actstr = self.act.__class__.__name__ if self.act is not None else 'No Activation'
|
|
s = ('{classname}(in_channels={in_channels}, out_channels={n_out_channels}')
|
|
s += (', ' + actstr)
|
|
if self.name is not None:
|
|
s += ', name=\'{name}\''
|
|
s += ')'
|
|
return s.format(classname=self.__class__.__name__, **self.__dict__)
|
|
|
|
def build(self, inputs_shape):
|
|
|
|
if inputs_shape is not None:
|
|
self.in_channels = inputs_shape[-1]
|
|
|
|
if self.in_channels / (self.scale**2) % 1 != 0:
|
|
raise Exception(
|
|
"SubpixelConv2d: The number of input channels == (scale x scale) x The number of output channels"
|
|
)
|
|
self.n_out_channels = int(self.in_channels / (self.scale**2))
|
|
self.depth_to_space = tl.ops.DepthToSpace(block_size=self.scale)
|
|
|
|
def forward(self, inputs):
|
|
if self._forward_state == False:
|
|
if self._built == False:
|
|
self.build(tl.get_tensor_shape(inputs))
|
|
self._built = True
|
|
self._forward_state = True
|
|
|
|
outputs = self._PS(X=inputs, r=self.scale, n_out_channels=self.n_out_channels)
|
|
if self.act is not None:
|
|
outputs = self.act(outputs)
|
|
return outputs
|
|
|
|
def _PS(self, X, r, n_out_channels):
|
|
|
|
_err_log = "SubpixelConv2d: The number of input channels == (scale x scale) x The number of output channels"
|
|
|
|
if n_out_channels >= 1:
|
|
if int(X.get_shape()[-1]) != (r**2) * n_out_channels:
|
|
raise Exception(_err_log)
|
|
|
|
X = self.depth_to_space(input=X)
|
|
else:
|
|
raise RuntimeError(_err_log)
|
|
|
|
return X
|