| |
| |
| |
| |
| |
| |
| |
| |
| |
| |
|
|
| import torch.nn as nn |
|
|
| from monai.networks.blocks.convolutions import Convolution |
| from monai.networks.blocks.upsample import UpSample |
| from monai.networks.layers.factories import Act, Norm |
|
|
|
|
| def get_norm_layer(spatial_dims: int, in_channels: int, norm_name: str, num_groups: int = 8): |
| if norm_name not in ["batch", "instance", "group"]: |
| raise ValueError(f"Unsupported normalization mode: {norm_name}") |
| else: |
| if norm_name == "group": |
| norm = Norm[norm_name](num_groups=num_groups, num_channels=in_channels) |
| else: |
| norm = Norm[norm_name, spatial_dims](in_channels) |
| if norm.bias is not None: |
| nn.init.zeros_(norm.bias) |
| if norm.weight is not None: |
| nn.init.ones_(norm.weight) |
| return norm |
|
|
|
|
| def get_conv_layer( |
| spatial_dims: int, in_channels: int, out_channels: int, kernel_size: int = 3, stride: int = 1, bias: bool = False |
| ): |
|
|
| return Convolution( |
| spatial_dims, |
| in_channels, |
| out_channels, |
| strides=stride, |
| kernel_size=kernel_size, |
| bias=bias, |
| conv_only=True, |
| ) |
|
|
|
|
| def get_upsample_layer(spatial_dims: int, in_channels: int, upsample_mode: str = "trilinear", scale_factor: int = 2): |
| up_module: nn.Module |
| if upsample_mode == "transpose": |
| up_module = UpSample( |
| spatial_dims, |
| in_channels, |
| scale_factor=scale_factor, |
| with_conv=True, |
| ) |
| else: |
| upsample_mode = "bilinear" if spatial_dims == 2 else "trilinear" |
| up_module = nn.Upsample(scale_factor=scale_factor, mode=upsample_mode, align_corners=False) |
| return up_module |
|
|
|
|
| class ResBlock(nn.Module): |
| """ |
| ResBlock employs skip connection and two convolution blocks and is used |
| in SegResNet based on `3D MRI brain tumor segmentation using autoencoder regularization |
| <https://arxiv.org/pdf/1810.11654.pdf>`_. |
| """ |
|
|
| def __init__( |
| self, |
| spatial_dims: int, |
| in_channels: int, |
| kernel_size: int = 3, |
| stride: int = 1, |
| bias: bool = False, |
| norm_name: str = "group", |
| num_groups: int = 8, |
| ) -> None: |
| """ |
| Args: |
| spatial_dims: number of spatial dimensions, could be 1, 2 or 3. |
| in_channels: number of input channels. |
| kernel_size: convolution kernel size, the value should be an odd number. Defaults to 3. |
| stride: convolution stride. Defaults to 1. |
| bias: whether to have a bias term in convolution layer. Defaults to ``True``. |
| norm_name: feature normalization type, this module only supports group norm, |
| batch norm and instance norm. Defaults to ``group``. |
| num_groups: number of groups to separate the channels into, in this module, |
| in_channels should be divisible by num_groups. Defaults to 8. |
| """ |
|
|
| super().__init__() |
|
|
| assert kernel_size % 2 == 1, "kernel_size should be an odd number." |
| assert in_channels % num_groups == 0, "in_channels should be divisible by num_groups." |
|
|
| self.norm1 = get_norm_layer(spatial_dims, in_channels, norm_name, num_groups=num_groups) |
| self.norm2 = get_norm_layer(spatial_dims, in_channels, norm_name, num_groups=num_groups) |
| self.relu = Act[Act.RELU](inplace=True) |
| self.conv1 = get_conv_layer(spatial_dims, in_channels, in_channels) |
| self.conv2 = get_conv_layer(spatial_dims, in_channels, in_channels) |
|
|
| def forward(self, x): |
|
|
| identity = x |
|
|
| x = self.norm1(x) |
| x = self.relu(x) |
| x = self.conv1(x) |
|
|
| x = self.norm2(x) |
| x = self.relu(x) |
| x = self.conv2(x) |
|
|
| x += identity |
|
|
| return x |
|
|