mmagic.models.editors.deblurganv2.deblurganv2_util

Module Contents

Classes

BasicConv2d	Base class for all neural network modules.
Mixed_5b	Base class for all neural network modules.
Block35	Base class for all neural network modules.
Mixed_6a	Base class for all neural network modules.
Block17	Base class for all neural network modules.
Mixed_7a	Base class for all neural network modules.
Block8	Base class for all neural network modules.
InceptionResNetV2	Define a inceptionresnetv2 model.
InvertedResidual	Base class for all neural network modules.
MobileNetV2	Base class for all neural network modules.

Functions

inceptionresnetv2([num_classes, pretrained])	return a inceptionresnetv2 network.
conv_bn(inp, oup, stride)
conv_1x1_bn(inp, oup)
get_norm_layer([norm_type])	Returns a norm layer of the specified type.

Attributes

pretrained_settings

mmagic.models.editors.deblurganv2.deblurganv2_util.pretrained_settings

class mmagic.models.editors.deblurganv2.deblurganv2_util.BasicConv2d(in_planes, out_planes, kernel_size, stride, padding=0)

Bases: torch.nn.Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call to(), etc.

备注

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

变量

training (bool) – Boolean represents whether this module is in training or evaluation mode.

forward(x)

Forward function.

参数

x (torch.Tensor) – You can directly input a torch.Tensor.

返回

torch.tensor will be returned.

返回类型

torch.Tensor

class mmagic.models.editors.deblurganv2.deblurganv2_util.Mixed_5b

Bases: torch.nn.Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call to(), etc.

备注

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

变量

training (bool) – Boolean represents whether this module is in training or evaluation mode.

forward(x)

Forward function.

参数

x (torch.Tensor) – You can directly input a torch.Tensor.

返回

torch.tensor will be returned.

返回类型

torch.Tensor

class mmagic.models.editors.deblurganv2.deblurganv2_util.Block35(scale=1.0)

Bases: torch.nn.Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call to(), etc.

备注

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

变量

training (bool) – Boolean represents whether this module is in training or evaluation mode.

forward(x)

Forward function.

参数

x (torch.Tensor) – You can directly input a torch.Tensor.

返回

torch.tensor will be returned.

返回类型

torch.Tensor

class mmagic.models.editors.deblurganv2.deblurganv2_util.Mixed_6a

Bases: torch.nn.Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call to(), etc.

备注

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

变量

training (bool) – Boolean represents whether this module is in training or evaluation mode.

forward(x)

Forward function.

参数

x (torch.Tensor) – You can directly input a torch.Tensor.

返回

torch.tensor will be returned.

返回类型

torch.Tensor

class mmagic.models.editors.deblurganv2.deblurganv2_util.Block17(scale=1.0)

Bases: torch.nn.Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call to(), etc.

备注

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

变量

training (bool) – Boolean represents whether this module is in training or evaluation mode.

forward(x)

Forward function.

参数

x (torch.Tensor) – You can directly input a torch.Tensor.

返回

torch.tensor will be returned.

返回类型

torch.Tensor

class mmagic.models.editors.deblurganv2.deblurganv2_util.Mixed_7a

Bases: torch.nn.Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call to(), etc.

备注

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

变量

training (bool) – Boolean represents whether this module is in training or evaluation mode.

forward(x)

Forward function.

参数

x (torch.Tensor) – You can directly input a torch.Tensor.

返回

torch.tensor will be returned.

返回类型

torch.Tensor

class mmagic.models.editors.deblurganv2.deblurganv2_util.Block8(scale=1.0, noReLU=False)

Bases: torch.nn.Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call to(), etc.

备注

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

变量

training (bool) – Boolean represents whether this module is in training or evaluation mode.

forward(x)

Forward function.

参数

x (torch.Tensor) – You can directly input a torch.Tensor.

返回

torch.tensor will be returned.

返回类型

torch.Tensor

class mmagic.models.editors.deblurganv2.deblurganv2_util.InceptionResNetV2(num_classes=1001)

Bases: torch.nn.Module

Define a inceptionresnetv2 model.

features(input)

Get network features.

参数

input (torch.tensor) – You can directly input a torch.Tensor.

logits(features)

Get features logits.

参数

features (torch.tensor) – You can directly input a torch.Tensor.

forward(input)

Forward function.

参数

input (torch.Tensor) – You can directly input a torch.Tensor.

返回

torch.tensor will be returned.

返回类型

torch.Tensor

mmagic.models.editors.deblurganv2.deblurganv2_util.inceptionresnetv2(num_classes=1000, pretrained='imagenet')

return a inceptionresnetv2 network.

mmagic.models.editors.deblurganv2.deblurganv2_util.conv_bn(inp, oup, stride)

mmagic.models.editors.deblurganv2.deblurganv2_util.conv_1x1_bn(inp, oup)

class mmagic.models.editors.deblurganv2.deblurganv2_util.InvertedResidual(inp, oup, stride, expand_ratio)

Bases: torch.nn.Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call to(), etc.

备注

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

变量

training (bool) – Boolean represents whether this module is in training or evaluation mode.

forward(x)

Forward function.

参数

x (torch.Tensor) – You can directly input a torch.Tensor.

返回

torch.tensor will be returned.

返回类型

torch.Tensor

class mmagic.models.editors.deblurganv2.deblurganv2_util.MobileNetV2(n_class=1000, input_size=224, width_mult=1.0)

Bases: torch.nn.Module

Base class for all neural network modules.

Your models should also subclass this class.

Modules can also contain other Modules, allowing to nest them in a tree structure. You can assign the submodules as regular attributes:

import torch.nn as nn
import torch.nn.functional as F

class Model(nn.Module):
    def __init__(self):
        super().__init__()
        self.conv1 = nn.Conv2d(1, 20, 5)
        self.conv2 = nn.Conv2d(20, 20, 5)

    def forward(self, x):
        x = F.relu(self.conv1(x))
        return F.relu(self.conv2(x))

Submodules assigned in this way will be registered, and will have their parameters converted too when you call to(), etc.

备注

As per the example above, an __init__() call to the parent class must be made before assignment on the child.

变量

training (bool) – Boolean represents whether this module is in training or evaluation mode.

forward(x)

Forward function.

参数

x (torch.Tensor) – You can directly input a torch.Tensor.

返回

torch.tensor will be returned.

返回类型

torch.Tensor

_initialize_weights()

mmagic.models.editors.deblurganv2.deblurganv2_util.get_norm_layer(norm_type='instance')

Returns a norm layer of the specified type.

参数

norm_type (Str) – norm layer type