Module `heyvi.model.ResNets_3D_PyTorch.resnet`

Expand source code Browse git

import math
from functools import partial

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


def get_inplanes():
    return [64, 128, 256, 512]


def conv3x3x3(in_planes, out_planes, stride=1):
    return nn.Conv3d(in_planes,
                     out_planes,
                     kernel_size=3,
                     stride=stride,
                     padding=1,
                     bias=False)


def conv1x1x1(in_planes, out_planes, stride=1):
    return nn.Conv3d(in_planes,
                     out_planes,
                     kernel_size=1,
                     stride=stride,
                     bias=False)


class BasicBlock(nn.Module):
    expansion = 1

    def __init__(self, in_planes, planes, stride=1, downsample=None):
        super().__init__()

        self.conv1 = conv3x3x3(in_planes, planes, stride)
        self.bn1 = nn.BatchNorm3d(planes)
        self.relu = nn.ReLU(inplace=True)
        self.conv2 = conv3x3x3(planes, planes)
        self.bn2 = nn.BatchNorm3d(planes)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x):
        residual = x

        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)

        out = self.conv2(out)
        out = self.bn2(out)

        if self.downsample is not None:
            residual = self.downsample(x)

        out += residual
        out = self.relu(out)

        return out


class Bottleneck(nn.Module):
    expansion = 4

    def __init__(self, in_planes, planes, stride=1, downsample=None):
        super().__init__()

        self.conv1 = conv1x1x1(in_planes, planes)
        self.bn1 = nn.BatchNorm3d(planes)
        self.conv2 = conv3x3x3(planes, planes, stride)
        self.bn2 = nn.BatchNorm3d(planes)
        self.conv3 = conv1x1x1(planes, planes * self.expansion)
        self.bn3 = nn.BatchNorm3d(planes * self.expansion)
        self.relu = nn.ReLU(inplace=True)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x):
        residual = x

        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)

        out = self.conv2(out)
        out = self.bn2(out)
        out = self.relu(out)

        out = self.conv3(out)
        out = self.bn3(out)

        if self.downsample is not None:
            residual = self.downsample(x)

        out += residual
        out = self.relu(out)

        return out


class ResNet(nn.Module):

    def __init__(self,
                 block,
                 layers,
                 block_inplanes,
                 n_input_channels=3,
                 conv1_t_size=7,
                 conv1_t_stride=1,
                 no_max_pool=False,
                 shortcut_type='B',
                 widen_factor=1.0,
                 unitnorm=False,
                 n_classes=400):
        super().__init__()

        block_inplanes = [int(x * widen_factor) for x in block_inplanes]

        self.in_planes = block_inplanes[0]
        self.no_max_pool = no_max_pool
        self._unitnorm = unitnorm  # embedding layer

        self.conv1 = nn.Conv3d(n_input_channels,
                               self.in_planes,
                               kernel_size=(conv1_t_size, 7, 7),
                               stride=(conv1_t_stride, 2, 2),
                               padding=(conv1_t_size // 2, 3, 3),
                               bias=False)
        self.bn1 = nn.BatchNorm3d(self.in_planes)
        self.relu = nn.ReLU(inplace=True)
        self.maxpool = nn.MaxPool3d(kernel_size=3, stride=2, padding=1)
        self.layer1 = self._make_layer(block, block_inplanes[0], layers[0],
                                       shortcut_type)
        self.layer2 = self._make_layer(block,
                                       block_inplanes[1],
                                       layers[1],
                                       shortcut_type,
                                       stride=2)
        self.layer3 = self._make_layer(block,
                                       block_inplanes[2],
                                       layers[2],
                                       shortcut_type,
                                       stride=2)
        self.layer4 = self._make_layer(block,
                                       block_inplanes[3],
                                       layers[3],
                                       shortcut_type,
                                       stride=2)

        self.avgpool = nn.AdaptiveAvgPool3d((1, 1, 1))
        self.fc = nn.Linear(block_inplanes[3] * block.expansion, n_classes)

        for m in self.modules():
            if isinstance(m, nn.Conv3d):
                nn.init.kaiming_normal_(m.weight,
                                        mode='fan_out',
                                        nonlinearity='relu')
            elif isinstance(m, nn.BatchNorm3d):
                nn.init.constant_(m.weight, 1)
                nn.init.constant_(m.bias, 0)

    def _downsample_basic_block(self, x, planes, stride):
        out = F.avg_pool3d(x, kernel_size=1, stride=stride)
        zero_pads = torch.zeros(out.size(0), planes - out.size(1), out.size(2),
                                out.size(3), out.size(4))
        if isinstance(out.data, torch.cuda.FloatTensor):
            zero_pads = zero_pads.cuda()

        out = torch.cat([out.data, zero_pads], dim=1)

        return out

    def _make_layer(self, block, planes, blocks, shortcut_type, stride=1):
        downsample = None
        if stride != 1 or self.in_planes != planes * block.expansion:
            if shortcut_type == 'A':
                downsample = partial(self._downsample_basic_block,
                                     planes=planes * block.expansion,
                                     stride=stride)
            else:
                downsample = nn.Sequential(
                    conv1x1x1(self.in_planes, planes * block.expansion, stride),
                    nn.BatchNorm3d(planes * block.expansion))

        layers = []
        layers.append(
            block(in_planes=self.in_planes,
                  planes=planes,
                  stride=stride,
                  downsample=downsample))
        self.in_planes = planes * block.expansion
        for i in range(1, blocks):
            layers.append(block(self.in_planes, planes))

        return nn.Sequential(*layers)

    def forward(self, x):
        x = self.conv1(x)
        x = self.bn1(x)
        x = self.relu(x)
        if not self.no_max_pool:
            x = self.maxpool(x)

        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        x = self.layer4(x)

        x = self.avgpool(x)

        x = x.view(x.size(0), -1)
        
        if self._unitnorm:
            x = F.normalize(x, p=2)

        x = self.fc(x)
        return x


def generate_model(model_depth, **kwargs):
    assert model_depth in [10, 18, 34, 50, 101, 152, 200]

    if model_depth == 10:
        model = ResNet(BasicBlock, [1, 1, 1, 1], get_inplanes(), **kwargs)
    elif model_depth == 18:
        model = ResNet(BasicBlock, [2, 2, 2, 2], get_inplanes(), **kwargs)
    elif model_depth == 34:
        model = ResNet(BasicBlock, [3, 4, 6, 3], get_inplanes(), **kwargs)
    elif model_depth == 50:
        model = ResNet(Bottleneck, [3, 4, 6, 3], get_inplanes(), **kwargs)
    elif model_depth == 101:
        model = ResNet(Bottleneck, [3, 4, 23, 3], get_inplanes(), **kwargs)
    elif model_depth == 152:
        model = ResNet(Bottleneck, [3, 8, 36, 3], get_inplanes(), **kwargs)
    elif model_depth == 200:
        model = ResNet(Bottleneck, [3, 24, 36, 3], get_inplanes(), **kwargs)

    return model

Functions

def conv1x1x1(in_planes, out_planes, stride=1)

Expand source code Browse git

def conv1x1x1(in_planes, out_planes, stride=1):
    return nn.Conv3d(in_planes,
                     out_planes,
                     kernel_size=1,
                     stride=stride,
                     bias=False)

def conv3x3x3(in_planes, out_planes, stride=1)

Expand source code Browse git

def conv3x3x3(in_planes, out_planes, stride=1):
    return nn.Conv3d(in_planes,
                     out_planes,
                     kernel_size=3,
                     stride=stride,
                     padding=1,
                     bias=False)

def generate_model(model_depth, **kwargs)

Expand source code Browse git

def generate_model(model_depth, **kwargs):
    assert model_depth in [10, 18, 34, 50, 101, 152, 200]

    if model_depth == 10:
        model = ResNet(BasicBlock, [1, 1, 1, 1], get_inplanes(), **kwargs)
    elif model_depth == 18:
        model = ResNet(BasicBlock, [2, 2, 2, 2], get_inplanes(), **kwargs)
    elif model_depth == 34:
        model = ResNet(BasicBlock, [3, 4, 6, 3], get_inplanes(), **kwargs)
    elif model_depth == 50:
        model = ResNet(Bottleneck, [3, 4, 6, 3], get_inplanes(), **kwargs)
    elif model_depth == 101:
        model = ResNet(Bottleneck, [3, 4, 23, 3], get_inplanes(), **kwargs)
    elif model_depth == 152:
        model = ResNet(Bottleneck, [3, 8, 36, 3], get_inplanes(), **kwargs)
    elif model_depth == 200:
        model = ResNet(Bottleneck, [3, 24, 36, 3], get_inplanes(), **kwargs)

    return model

def get_inplanes()

Expand source code Browse git

def get_inplanes():
    return [64, 128, 256, 512]

Classes

class BasicBlock (in_planes, planes, stride=1, downsample=None)

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 :meth:to, etc.

Note

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

:ivar training: Boolean represents whether this module is in training or evaluation mode. :vartype training: bool

Initializes internal Module state, shared by both nn.Module and ScriptModule.

Expand source code Browse git

class BasicBlock(nn.Module):
    expansion = 1

    def __init__(self, in_planes, planes, stride=1, downsample=None):
        super().__init__()

        self.conv1 = conv3x3x3(in_planes, planes, stride)
        self.bn1 = nn.BatchNorm3d(planes)
        self.relu = nn.ReLU(inplace=True)
        self.conv2 = conv3x3x3(planes, planes)
        self.bn2 = nn.BatchNorm3d(planes)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x):
        residual = x

        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)

        out = self.conv2(out)
        out = self.bn2(out)

        if self.downsample is not None:
            residual = self.downsample(x)

        out += residual
        out = self.relu(out)

        return out

Ancestors

torch.nn.modules.module.Module

Class variables

var expansion

Methods

def forward(self, x) ‑> Callable[..., Any]

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Although the recipe for forward pass needs to be defined within this function, one should call the :class:Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

Expand source code Browse git

def forward(self, x):
    residual = x

    out = self.conv1(x)
    out = self.bn1(out)
    out = self.relu(out)

    out = self.conv2(out)
    out = self.bn2(out)

    if self.downsample is not None:
        residual = self.downsample(x)

    out += residual
    out = self.relu(out)

    return out

class Bottleneck (in_planes, planes, stride=1, downsample=None)

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 :meth:to, etc.

Note

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

:ivar training: Boolean represents whether this module is in training or evaluation mode. :vartype training: bool

Initializes internal Module state, shared by both nn.Module and ScriptModule.

Expand source code Browse git

class Bottleneck(nn.Module):
    expansion = 4

    def __init__(self, in_planes, planes, stride=1, downsample=None):
        super().__init__()

        self.conv1 = conv1x1x1(in_planes, planes)
        self.bn1 = nn.BatchNorm3d(planes)
        self.conv2 = conv3x3x3(planes, planes, stride)
        self.bn2 = nn.BatchNorm3d(planes)
        self.conv3 = conv1x1x1(planes, planes * self.expansion)
        self.bn3 = nn.BatchNorm3d(planes * self.expansion)
        self.relu = nn.ReLU(inplace=True)
        self.downsample = downsample
        self.stride = stride

    def forward(self, x):
        residual = x

        out = self.conv1(x)
        out = self.bn1(out)
        out = self.relu(out)

        out = self.conv2(out)
        out = self.bn2(out)
        out = self.relu(out)

        out = self.conv3(out)
        out = self.bn3(out)

        if self.downsample is not None:
            residual = self.downsample(x)

        out += residual
        out = self.relu(out)

        return out

Ancestors

torch.nn.modules.module.Module

Class variables

var expansion

Methods

def forward(self, x) ‑> Callable[..., Any]

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Expand source code Browse git

def forward(self, x):
    residual = x

    out = self.conv1(x)
    out = self.bn1(out)
    out = self.relu(out)

    out = self.conv2(out)
    out = self.bn2(out)
    out = self.relu(out)

    out = self.conv3(out)
    out = self.bn3(out)

    if self.downsample is not None:
        residual = self.downsample(x)

    out += residual
    out = self.relu(out)

    return out

class ResNet (block, layers, block_inplanes, n_input_channels=3, conv1_t_size=7, conv1_t_stride=1, no_max_pool=False, shortcut_type='B', widen_factor=1.0, unitnorm=False, n_classes=400)

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 :meth:to, etc.

Note

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

:ivar training: Boolean represents whether this module is in training or evaluation mode. :vartype training: bool

Initializes internal Module state, shared by both nn.Module and ScriptModule.

Expand source code Browse git

class ResNet(nn.Module):

    def __init__(self,
                 block,
                 layers,
                 block_inplanes,
                 n_input_channels=3,
                 conv1_t_size=7,
                 conv1_t_stride=1,
                 no_max_pool=False,
                 shortcut_type='B',
                 widen_factor=1.0,
                 unitnorm=False,
                 n_classes=400):
        super().__init__()

        block_inplanes = [int(x * widen_factor) for x in block_inplanes]

        self.in_planes = block_inplanes[0]
        self.no_max_pool = no_max_pool
        self._unitnorm = unitnorm  # embedding layer

        self.conv1 = nn.Conv3d(n_input_channels,
                               self.in_planes,
                               kernel_size=(conv1_t_size, 7, 7),
                               stride=(conv1_t_stride, 2, 2),
                               padding=(conv1_t_size // 2, 3, 3),
                               bias=False)
        self.bn1 = nn.BatchNorm3d(self.in_planes)
        self.relu = nn.ReLU(inplace=True)
        self.maxpool = nn.MaxPool3d(kernel_size=3, stride=2, padding=1)
        self.layer1 = self._make_layer(block, block_inplanes[0], layers[0],
                                       shortcut_type)
        self.layer2 = self._make_layer(block,
                                       block_inplanes[1],
                                       layers[1],
                                       shortcut_type,
                                       stride=2)
        self.layer3 = self._make_layer(block,
                                       block_inplanes[2],
                                       layers[2],
                                       shortcut_type,
                                       stride=2)
        self.layer4 = self._make_layer(block,
                                       block_inplanes[3],
                                       layers[3],
                                       shortcut_type,
                                       stride=2)

        self.avgpool = nn.AdaptiveAvgPool3d((1, 1, 1))
        self.fc = nn.Linear(block_inplanes[3] * block.expansion, n_classes)

        for m in self.modules():
            if isinstance(m, nn.Conv3d):
                nn.init.kaiming_normal_(m.weight,
                                        mode='fan_out',
                                        nonlinearity='relu')
            elif isinstance(m, nn.BatchNorm3d):
                nn.init.constant_(m.weight, 1)
                nn.init.constant_(m.bias, 0)

    def _downsample_basic_block(self, x, planes, stride):
        out = F.avg_pool3d(x, kernel_size=1, stride=stride)
        zero_pads = torch.zeros(out.size(0), planes - out.size(1), out.size(2),
                                out.size(3), out.size(4))
        if isinstance(out.data, torch.cuda.FloatTensor):
            zero_pads = zero_pads.cuda()

        out = torch.cat([out.data, zero_pads], dim=1)

        return out

    def _make_layer(self, block, planes, blocks, shortcut_type, stride=1):
        downsample = None
        if stride != 1 or self.in_planes != planes * block.expansion:
            if shortcut_type == 'A':
                downsample = partial(self._downsample_basic_block,
                                     planes=planes * block.expansion,
                                     stride=stride)
            else:
                downsample = nn.Sequential(
                    conv1x1x1(self.in_planes, planes * block.expansion, stride),
                    nn.BatchNorm3d(planes * block.expansion))

        layers = []
        layers.append(
            block(in_planes=self.in_planes,
                  planes=planes,
                  stride=stride,
                  downsample=downsample))
        self.in_planes = planes * block.expansion
        for i in range(1, blocks):
            layers.append(block(self.in_planes, planes))

        return nn.Sequential(*layers)

    def forward(self, x):
        x = self.conv1(x)
        x = self.bn1(x)
        x = self.relu(x)
        if not self.no_max_pool:
            x = self.maxpool(x)

        x = self.layer1(x)
        x = self.layer2(x)
        x = self.layer3(x)
        x = self.layer4(x)

        x = self.avgpool(x)

        x = x.view(x.size(0), -1)
        
        if self._unitnorm:
            x = F.normalize(x, p=2)

        x = self.fc(x)
        return x

Ancestors

torch.nn.modules.module.Module

Methods

def forward(self, x) ‑> Callable[..., Any]

Defines the computation performed at every call.

Should be overridden by all subclasses.

Note

Expand source code Browse git

def forward(self, x):
    x = self.conv1(x)
    x = self.bn1(x)
    x = self.relu(x)
    if not self.no_max_pool:
        x = self.maxpool(x)

    x = self.layer1(x)
    x = self.layer2(x)
    x = self.layer3(x)
    x = self.layer4(x)

    x = self.avgpool(x)

    x = x.view(x.size(0), -1)
    
    if self._unitnorm:
        x = F.normalize(x, p=2)

    x = self.fc(x)
    return x