Network architectures

Blocks

Convolution

class monai.networks.blocks.Convolution(dimensions, in_channels, out_channels, strides=1, kernel_size=3, act='PRELU', norm='INSTANCE', dropout=None, dilation=1, bias=True, conv_only=False, is_transposed=False)

Constructs a convolution with optional dropout, normalization, and activation layers.

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

ResidualUnit

class monai.networks.blocks.ResidualUnit(dimensions, in_channels, out_channels, strides=1, kernel_size=3, subunits=2, act='PRELU', norm='INSTANCE', dropout=None, dilation=1, bias=True, last_conv_only=False)

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

forward(x)

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 Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

Squeeze-and-Excitation

class monai.networks.blocks.ChannelSELayer(spatial_dims, in_channels, r=2, acti_type_1='relu', acti_type_2='sigmoid')

Re-implementation of the Squeeze-and-Excitation block based on: “Hu et al., Squeeze-and-Excitation Networks, https://arxiv.org/abs/1709.01507”.

Parameters

• spatial_dims (int) – number of spatial dimensions, could be 1, 2, or 3.

• in_channels (int) – number of input channels.

• r (int) – the reduction ratio r in the paper. Defaults to 2.

• acti_type_1 (str) – activation type of the hidden squeeze layer. Defaults to “relu”.

• acti_type_2 (str) – activation type of the output squeeze layer. Defaults to “sigmoid”.

Raises

ValueError – r must be a positive number smaller than in_channels.

forward(x)

Parameters

x (Tensor) – in shape (batch, channel, spatial_1[, spatial_2, …]).

Residual Squeeze-and-Excitation

class monai.networks.blocks.ResidualSELayer(spatial_dims, in_channels, r=2, acti_type_1='leakyrelu', acti_type_2='relu')

A “squeeze-and-excitation”-like layer with a residual connection.

Parameters

• spatial_dims (int) – number of spatial dimensions, could be 1, 2, or 3.

• in_channels (int) – number of input channels.

• r (int) – the reduction ratio r in the paper. Defaults to 2.

• acti_type_1 (str) – defaults to “leakyrelu”.

• acti_type_2 (str) – defaults to “relu”.

See also :monai.networks.blocks.ChannelSELayer.

forward(x)

Parameters

x (Tensor) – in shape (batch, channel, spatial_1[, spatial_2, …]).

Simple ASPP

class monai.networks.blocks.SimpleASPP(spatial_dims, in_channels, conv_out_channels, kernel_sizes=(1, 3, 3, 3), dilations=(1, 2, 4, 6), norm_type='BATCH', acti_type='LEAKYRELU')

A simplified version of the atrous spatial pyramid pooling (ASPP) module.

Chen et al., Encoder-Decoder with Atrous Separable Convolution for Semantic Image Segmentation. https://arxiv.org/abs/1802.02611

Wang et al., A Noise-robust Framework for Automatic Segmentation of COVID-19 Pneumonia Lesions from CT Images. https://ieeexplore.ieee.org/document/9109297

Parameters

• spatial_dims (int) – number of spatial dimensions, could be 1, 2, or 3.

• in_channels (int) – number of input channels.

• conv_out_channels (int) – number of output channels of each atrous conv. The final number of output channels is conv_out_channels * len(kernel_sizes).

• kernel_sizes – a sequence of four convolutional kernel sizes. Defaults to (1, 3, 3, 3) for four (dilated) convolutions.

• dilations – a sequence of four convolutional dilation parameters. Defaults to (1, 2, 4, 6) for four (dilated) convolutions.

• norm_type – final kernel-size-one convolution normalization type. Defaults to batch norm.

• acti_type – final kernel-size-one convolution activation type. Defaults to leaky ReLU.

Raises

ValueError – len(kernel_sizes) and len(dilations) must be the same.

forward(x)

Parameters

x (Tensor) – in shape (batch, channel, spatial_1[, spatial_2, …]).

Return type

Tensor

MaxAvgPooling

class monai.networks.blocks.MaxAvgPool(spatial_dims, kernel_size, stride=None, padding=0, ceil_mode=False)

Downsample with both maxpooling and avgpooling, double the channel size by concatenating the downsampled feature maps.

Parameters

• spatial_dims (int) – number of spatial dimensions of the input image.

• kernel_size – the kernel size of both pooling operations.

• stride – the stride of the window. Default value is kernel_size.

• padding – implicit zero padding to be added to both pooling operations.

• ceil_mode (bool) – when True, will use ceil instead of floor to compute the output shape.

forward(x)

Parameters

x (Tensor) – Tensor in shape (batch, channel, spatial_1[, spatial_2, …]).

Return type

Tensor

Returns

Tensor in shape (batch, 2*channel, spatial_1[, spatial_2, …]).

Upsampling

class monai.networks.blocks.UpSample(spatial_dims, in_channels, out_channels=None, scale_factor=2, with_conv=False, mode=<UpsampleMode.LINEAR: 'linear'>, align_corners=True)

Upsample with either kernel 1 conv + interpolation or transposed conv.

Parameters

• spatial_dims (int) – number of spatial dimensions of the input image.

• in_channels (int) – number of channels of the input image.

• out_channels (Optional[int]) – number of channels of the output image. Defaults to in_channels.

• scale_factor – multiplier for spatial size. Has to match input size if it is a tuple. Defaults to 2.

• with_conv (bool) – whether to use a transposed convolution for upsampling. Defaults to False.

• mode (Union[UpsampleMode, str]) – {"nearest", "linear", "bilinear", "bicubic", "trilinear"} If ends with "linear" will use spatial dims to determine the correct interpolation. This corresponds to linear, bilinear, trilinear for 1D, 2D, and 3D respectively. The interpolation mode. Defaults to "linear". See also: https://pytorch.org/docs/stable/nn.html#upsample

• align_corners (Optional[bool]) – set the align_corners parameter of torch.nn.Upsample. Defaults to True.

forward(x)

Parameters

x (Tensor) – Tensor in shape (batch, channel, spatial_1[, spatial_2, …).

Return type

Tensor

Layers

Factories

Defines factories for creating layers in generic, extensible, and dimensionally independent ways. A separate factory object is created for each type of layer, and factory functions keyed to names are added to these objects. Whenever a layer is requested the factory name and any necessary arguments are passed to the factory object. The return value is typically a type but can be any callable producing a layer object.

The factory objects contain functions keyed to names converted to upper case, these names can be referred to as members of the factory so that they can function as constant identifiers. eg. instance normalisation is named Norm.INSTANCE.

For example, to get a transpose convolution layer the name is needed and then a dimension argument is provided which is passed to the factory function:

dimension = 3
name = Conv.CONVTRANS
conv = Conv[name, dimension]

This allows the dimension value to be set in the constructor, for example so that the dimensionality of a network is parameterizable. Not all factories require arguments after the name, the caller must be aware which are required.

Defining new factories involves creating the object then associating it with factory functions:

fact = LayerFactory()

@fact.factory_function('test')
def make_something(x, y):
    # do something with x and y to choose which layer type to return
    return SomeLayerType
...

# request object from factory TEST with 1 and 2 as values for x and y
layer = fact[fact.TEST, 1, 2]

Typically the caller of a factory would know what arguments to pass (ie. the dimensionality of the requested type) but can be parameterized with the factory name and the arguments to pass to the created type at instantiation time:

def use_factory(fact_args):
    fact_name, type_args = split_args
    layer_type = fact[fact_name, 1, 2]
    return layer_type(**type_args)
...

kw_args = {'arg0':0, 'arg1':True}
layer = use_factory( (fact.TEST, kwargs) )

class monai.networks.layers.factories.LayerFactory

Factory object for creating layers, this uses given factory functions to actually produce the types or constructing callables. These functions are referred to by name and can be added at any time.

add_factory_callable(name, func)

Add the factory function to this object under the given name.

factory_function(name)

Decorator for adding a factory function with the given name.

get_constructor(factory_name, *args)

Get the constructor for the given factory name and arguments.

Raises

ValueError – Factories must be selected by name

Return type

Any

property names

Produces all factory names.

SkipConnection

class monai.networks.layers.SkipConnection(submodule, cat_dim=1)

Concats the forward pass input with the result from the given submodule.

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

forward(x)

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 Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

Flatten

class monai.networks.layers.Flatten

Flattens the given input in the forward pass to be [B,-1] in shape.

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

forward(x)

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 Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

GaussianFilter

class monai.networks.layers.GaussianFilter(spatial_dims, sigma, truncated=4.0)

Parameters

• spatial_dims (int) – number of spatial dimensions of the input image. must have shape (Batch, channels, H[, W, …]).

• sigma (float or sequence of floats) – std.

• truncated (float) – spreads how many stds.

forward(x)

Parameters

x (tensor) – in shape [Batch, chns, H, W, D].

Raises

TypeError – x must be a Tensor, got {type(x).__name__}.

Affine Transform

class monai.networks.layers.AffineTransform(spatial_size=None, normalized=False, mode=<GridSampleMode.BILINEAR: 'bilinear'>, padding_mode=<GridSamplePadMode.ZEROS: 'zeros'>, align_corners=False, reverse_indexing=True)

Apply affine transformations with a batch of affine matrices.

When normalized=False and reverse_indexing=True, it does the commonly used resampling in the ‘pull’ direction following the scipy.ndimage.affine_transform convention. In this case theta is equivalent to (ndim+1, ndim+1) input matrix of scipy.ndimage.affine_transform, operates on homogeneous coordinates. See also: https://docs.scipy.org/doc/scipy/reference/generated/scipy.ndimage.affine_transform.html

When normalized=True and reverse_indexing=False, it applies theta to the normalized coordinates (coords. in the range of [-1, 1]) directly. This is often used with align_corners=False to achieve resolution-agnostic resampling, thus useful as a part of trainable modules such as the spatial transformer networks. See also: https://pytorch.org/tutorials/intermediate/spatial_transformer_tutorial.html

Parameters

• spatial_size (list or tuple of int) – output spatial shape, the full output shape will be [N, C, *spatial_size] where N and C are inferred from the src input of self.forward.

• normalized (bool) – indicating whether the provided affine matrix theta is defined for the normalized coordinates. If normalized=False, theta will be converted to operate on normalized coordinates as pytorch affine_grid works with the normalized coordinates.

• mode (Union[GridSampleMode, str]) – {"bilinear", "nearest"} Interpolation mode to calculate output values. Defaults to "bilinear". See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample

• padding_mode (Union[GridSamplePadMode, str]) – {"zeros", "border", "reflection"} Padding mode for outside grid values. Defaults to "zeros". See also: https://pytorch.org/docs/stable/nn.functional.html#grid-sample

• align_corners (bool) – see also https://pytorch.org/docs/stable/nn.functional.html#grid-sample.

• reverse_indexing (bool) – whether to reverse the spatial indexing of image and coordinates. set to False if theta follows pytorch’s default “D, H, W” convention. set to True if theta follows scipy.ndimage default “i, j, k” convention.

forward(src, theta, spatial_size=None)

theta must be an affine transformation matrix with shape 3x3 or Nx3x3 or Nx2x3 or 2x3 for spatial 2D transforms, 4x4 or Nx4x4 or Nx3x4 or 3x4 for spatial 3D transforms, where N is the batch size. theta will be converted into float Tensor for the computation.

Parameters

• src (array_like) – image in spatial 2D or 3D (N, C, spatial_dims), where N is the batch dim, C is the number of channels.

• theta (array_like) – Nx3x3, Nx2x3, 3x3, 2x3 for spatial 2D inputs, Nx4x4, Nx3x4, 3x4, 4x4 for spatial 3D inputs. When the batch dimension is omitted, theta will be repeated N times, N is the batch dim of src.

• spatial_size (list or tuple of int) – output spatial shape, the full output shape will be [N, C, *spatial_size] where N and C are inferred from the src.

Raises

• TypeError – both src and theta must be torch Tensor, got {type(src).__name__}, {type(theta).__name__}.

• ValueError – affine must be Nxdxd or dxd.

• ValueError – affine must be Nx3x3 or Nx4x4, got: {theta.shape}.

• ValueError – src must be spatially 2D or 3D.

• ValueError – batch dimension of affine and image does not match, got affine: {} and image: {}.

Utilities

monai.networks.layers.convutils.same_padding(kernel_size, dilation=1)

Return the padding value needed to ensure a convolution using the given kernel size produces an output of the same shape as the input for a stride of 1, otherwise ensure a shape of the input divided by the stride rounded down.

Raises

NotImplementedError – same padding not available for k={kernel_size} and d={dilation}.

monai.networks.layers.convutils.calculate_out_shape(in_shape, kernel_size, stride, padding)

Calculate the output tensor shape when applying a convolution to a tensor of shape inShape with kernel size kernel_size, stride value stride, and input padding value padding. All arguments can be scalars or multiple values, return value is a scalar if all inputs are scalars.

monai.networks.layers.convutils.gaussian_1d(sigma, truncated=4.0)

one dimensional gaussian kernel.

Parameters

• sigma – std of the kernel

• truncated – tail length

Returns

1D numpy array

Raises

ValueError – sigma must be positive

Nets

Densenet3D

class monai.networks.nets.DenseNet(spatial_dims, in_channels, out_channels, init_features=64, growth_rate=32, block_config=(6, 12, 24, 16), bn_size=4, dropout_prob=0.0)

Densenet based on: “Densely Connected Convolutional Networks” https://arxiv.org/pdf/1608.06993.pdf Adapted from PyTorch Hub 2D version: https://github.com/pytorch/vision/blob/master/torchvision/models/densenet.py

Parameters

• spatial_dims (int) – number of spatial dimensions of the input image.

• in_channels (int) – number of the input channel.

• out_channels (int) – number of the output classes.

• init_features (int) – number of filters in the first convolution layer.

• growth_rate (int) – how many filters to add each layer (k in paper).

• block_config (tuple) – how many layers in each pooling block.

• bn_size (int) – multiplicative factor for number of bottle neck layers. (i.e. bn_size * k features in the bottleneck layer)

• dropout_prob (float) – dropout rate after each dense layer.

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

forward(x)

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 Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

monai.networks.nets.densenet121(**kwargs)

Return type

DenseNet

monai.networks.nets.densenet169(**kwargs)

Return type

DenseNet

monai.networks.nets.densenet201(**kwargs)

Return type

DenseNet

monai.networks.nets.densenet264(**kwargs)

Return type

DenseNet

Highresnet

class monai.networks.nets.HighResNet(spatial_dims=3, in_channels=1, out_channels=1, norm_type=<Normalisation.BATCH: 'batch'>, acti_type=<Activation.RELU: 'relu'>, dropout_prob=None, layer_params=({'name': 'conv_0', 'n_features': 16, 'kernel_size': 3}, {'name': 'res_1', 'n_features': 16, 'kernels': (3, 3), 'repeat': 3}, {'name': 'res_2', 'n_features': 32, 'kernels': (3, 3), 'repeat': 3}, {'name': 'res_3', 'n_features': 64, 'kernels': (3, 3), 'repeat': 3}, {'name': 'conv_1', 'n_features': 80, 'kernel_size': 1}, {'name': 'conv_2', 'kernel_size': 1}))

Reimplementation of highres3dnet based on Li et al., “On the compactness, efficiency, and representation of 3D convolutional networks: Brain parcellation as a pretext task”, IPMI ‘17

Adapted from: https://github.com/NifTK/NiftyNet/blob/v0.6.0/niftynet/network/highres3dnet.py https://github.com/fepegar/highresnet

Parameters

• spatial_dims (int) – number of spatial dimensions of the input image.

• in_channels (int) – number of input channels.

• out_channels (int) – number of output channels.

• norm_type (Union[Normalisation, str]) – {"batch", "instance"} Feature normalisation with batchnorm or instancenorm. Defaults to "batch".

• acti_type (Union[Activation, str]) – {"relu", "prelu", "relu6"} Non-linear activation using ReLU or PReLU. Defaults to "relu".

• dropout_prob (Optional[float]) – probability of the feature map to be zeroed (only applies to the penultimate conv layer).

• layer_params (a list of dictionaries) – specifying key parameters of each layer/block.

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

forward(x)

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 Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

class monai.networks.nets.HighResBlock(spatial_dims, in_channels, out_channels, kernels=(3, 3), dilation=1, norm_type=<Normalisation.INSTANCE: 'instance'>, acti_type=<Activation.RELU: 'relu'>, channel_matching=<ChannelMatching.PAD: 'pad'>)

Parameters

• kernels (list of int) – each integer k in kernels corresponds to a convolution layer with kernel size k.

• norm_type (Union[Normalisation, str]) – {"batch", "instance"} Feature normalisation with batchnorm or instancenorm. Defaults to "instance".

• acti_type (Union[Activation, str]) – {"relu", "prelu", "relu6"} Non-linear activation using ReLU or PReLU. Defaults to "relu".

• channel_matching (Union[ChannelMatching, str]) –

  {"pad", "project"} Specifies handling residual branch and conv branch channel mismatches. Defaults to "pad".

  • "pad": with zero padding.

  • "project": with a trainable conv with kernel size.

Raises

• ValueError – channel matching must be pad or project, got {channel_matching}.

• ValueError – in_channels > out_channels is incompatible with channel_matching=pad.

forward(x)

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 Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

Unet

class monai.networks.nets.UNet(dimensions, in_channels, out_channels, channels, strides, kernel_size=3, up_kernel_size=3, num_res_units=0, act='PRELU', norm='INSTANCE', dropout=0)

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

forward(x)

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 Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

monai.networks.nets.Unet

alias of monai.networks.nets.unet.UNet

monai.networks.nets.unet

alias of monai.networks.nets.unet.UNet

Generator

class monai.networks.nets.Generator(latent_shape, start_shape, channels, strides, kernel_size=3, num_res_units=2, act='PRELU', norm='INSTANCE', dropout=None, bias=True)

Defines a simple generator network accepting a latent vector and through a sequence of convolution layers constructs an output tensor of greater size and high dimensionality. The method _get_layer is used to create each of these layers, override this method to define layers beyond the default Convolution or ResidualUnit layers.

For example, a generator accepting a latent vector if shape (42,24) and producing an output volume of shape (1,64,64) can be constructed as:

gen = Generator((42, 24), (64, 8, 8), (32, 16, 1), (2, 2, 2))

Construct the generator network with the number of layers defined by channels and strides. In the forward pass a nn.Linear layer relates the input latent vector to a tensor of dimensions start_shape, this is then fed forward through the sequence of convolutional layers. The number of layers is defined by the length of channels and strides which must match, each layer having the number of output channels given in channels and an upsample factor given in strides (ie. a transpose convolution with that stride size).

Parameters

• latent_shape – tuple of integers stating the dimension of the input latent vector (minus batch dimension)

• start_shape – tuple of integers stating the dimension of the tensor to pass to convolution subnetwork

• channels – tuple of integers stating the output channels of each convolutional layer

• strides – tuple of integers stating the stride (upscale factor) of each convolutional layer

• kernel_size – integer or tuple of integers stating size of convolutional kernels

• num_res_units – integer stating number of convolutions in residual units, 0 means no residual units

• act – name or type defining activation layers

• norm – name or type defining normalization layers

• dropout – optional float value in range [0, 1] stating dropout probability for layers, None for no dropout

• bias – boolean stating if convolution layers should have a bias component

forward(x)

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 Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

Regressor

class monai.networks.nets.Regressor(in_shape, out_shape, channels, strides, kernel_size=3, num_res_units=2, act='PRELU', norm='INSTANCE', dropout=None, bias=True)

This defines a network for relating large-sized input tensors to small output tensors, ie. regressing large values to a prediction. An output of a single dimension can be used as value regression or multi-label classification prediction, an output of a single value can be used as a discriminator or critic prediction.

Construct the regressor network with the number of layers defined by channels and strides. Inputs are first passed through the convolutional layers in the forward pass, the output from this is then pass through a fully connected layer to relate them to the final output tensor.

Parameters

• in_shape – tuple of integers stating the dimension of the input tensor (minus batch dimension)

• out_shape – tuple of integers stating the dimension of the final output tensor

• channels – tuple of integers stating the output channels of each convolutional layer

• strides – tuple of integers stating the stride (downscale factor) of each convolutional layer

• kernel_size – integer or tuple of integers stating size of convolutional kernels

• num_res_units – integer stating number of convolutions in residual units, 0 means no residual units

• act – name or type defining activation layers

• norm – name or type defining normalization layers

• dropout – optional float value in range [0, 1] stating dropout probability for layers, None for no dropout

• bias – boolean stating if convolution layers should have a bias component

forward(x)

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 Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

Classifier

class monai.networks.nets.Classifier(in_shape, classes, channels, strides, kernel_size=3, num_res_units=2, act='PRELU', norm='INSTANCE', dropout=None, bias=True, last_act=None)

Defines a classification network from Regressor by specifying the output shape as a single dimensional tensor with size equal to the number of classes to predict. The final activation function can also be specified, eg. softmax or sigmoid.

Construct the regressor network with the number of layers defined by channels and strides. Inputs are first passed through the convolutional layers in the forward pass, the output from this is then pass through a fully connected layer to relate them to the final output tensor.

Parameters

• in_shape – tuple of integers stating the dimension of the input tensor (minus batch dimension)

• out_shape – tuple of integers stating the dimension of the final output tensor

• channels – tuple of integers stating the output channels of each convolutional layer

• strides – tuple of integers stating the stride (downscale factor) of each convolutional layer

• kernel_size – integer or tuple of integers stating size of convolutional kernels

• num_res_units – integer stating number of convolutions in residual units, 0 means no residual units

• act – name or type defining activation layers

• norm – name or type defining normalization layers

• dropout – optional float value in range [0, 1] stating dropout probability for layers, None for no dropout

• bias – boolean stating if convolution layers should have a bias component

Discriminator

class monai.networks.nets.Discriminator(in_shape, channels, strides, kernel_size=3, num_res_units=2, act='PRELU', norm='INSTANCE', dropout=0.25, bias=True, last_act='SIGMOID')

Defines a discriminator network from Classifier with a single output value and sigmoid activation by default. This is meant for use with GANs or other applications requiring a generic discriminator network.

Construct the regressor network with the number of layers defined by channels and strides. Inputs are first passed through the convolutional layers in the forward pass, the output from this is then pass through a fully connected layer to relate them to the final output tensor.

Parameters

• in_shape – tuple of integers stating the dimension of the input tensor (minus batch dimension)

• out_shape – tuple of integers stating the dimension of the final output tensor

• channels – tuple of integers stating the output channels of each convolutional layer

• strides – tuple of integers stating the stride (downscale factor) of each convolutional layer

• kernel_size – integer or tuple of integers stating size of convolutional kernels

• num_res_units – integer stating number of convolutions in residual units, 0 means no residual units

• act – name or type defining activation layers

• norm – name or type defining normalization layers

• dropout – optional float value in range [0, 1] stating dropout probability for layers, None for no dropout

• bias – boolean stating if convolution layers should have a bias component

Critic

class monai.networks.nets.Critic(in_shape, channels, strides, kernel_size=3, num_res_units=2, act='PRELU', norm='INSTANCE', dropout=0.25, bias=True)

Defines a critic network from Classifier with a single output value and no final activation. The final layer is nn.Flatten instead of nn.Linear, the final result is computed as the mean over the first dimension. This is meant to be used with Wassertein GANs.

Construct the regressor network with the number of layers defined by channels and strides. Inputs are first passed through the convolutional layers in the forward pass, the output from this is then pass through a fully connected layer to relate them to the final output tensor.

Parameters

• in_shape – tuple of integers stating the dimension of the input tensor (minus batch dimension)

• out_shape – tuple of integers stating the dimension of the final output tensor

• channels – tuple of integers stating the output channels of each convolutional layer

• strides – tuple of integers stating the stride (downscale factor) of each convolutional layer

• kernel_size – integer or tuple of integers stating size of convolutional kernels

• num_res_units – integer stating number of convolutions in residual units, 0 means no residual units

• act – name or type defining activation layers

• norm – name or type defining normalization layers

• dropout – optional float value in range [0, 1] stating dropout probability for layers, None for no dropout

• bias – boolean stating if convolution layers should have a bias component

forward(x)

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 Module instance afterwards instead of this since the former takes care of running the registered hooks while the latter silently ignores them.

Utilities

Utilities and types for defining networks, these depend on PyTorch.

monai.networks.utils.normal_init(m, std=0.02, normal_func=<function normal_>)

Initialize the weight and bias tensors of m’ and its submodules to values from a normal distribution with a stddev of `std’. Weight tensors of convolution and linear modules are initialized with a mean of 0, batch norm modules with a mean of 1. The callable `normal_func’, used to assign values, should have the same arguments as its default normal_(). This can be used with `nn.Module.apply to visit submodules of a network.

monai.networks.utils.normalize_transform(shape, device=None, dtype=None, align_corners=False)

Compute an affine matrix according to the input shape. The transform normalizes the homogeneous image coordinates to the range of [-1, 1].

Parameters

• shape (sequence of int) – input spatial shape

• device (torch device) – device on which the returned affine will be allocated.

• dtype (torch dtype) – data type of the returned affine

• align_corners (bool) – if True, consider -1 and 1 to refer to the centers of the corner pixels rather than the image corners. See also: https://pytorch.org/docs/stable/nn.functional.html#torch.nn.functional.grid_sample

monai.networks.utils.one_hot(labels, num_classes, dtype=torch.float32)

For a tensor labels of dimensions B1[spatial_dims], return a tensor of dimensions BN[spatial_dims] for num_classes N number of classes.

Example

For every value v = labels[b,1,h,w], the value in the result at [b,v,h,w] will be 1 and all others 0. Note that this will include the background label, thus a binary mask should be treated as having 2 classes.

monai.networks.utils.predict_segmentation(logits, mutually_exclusive=False, threshold=0.0)

Given the logits from a network, computing the segmentation by thresholding all values above 0 if multi-labels task, computing the argmax along the channel axis if multi-classes task, logits has shape BCHW[D].

Parameters

• logits (Tensor) – raw data of model output.

• mutually_exclusive (bool) – if True, logits will be converted into a binary matrix using a combination of argmax, which is suitable for multi-classes task. Defaults to False.

• threshold (float) – thresholding the prediction values if multi-labels task.

monai.networks.utils.to_norm_affine(affine, src_size, dst_size, align_corners=False)

Given affine defined for coordinates in the pixel space, compute the corresponding affine for the normalized coordinates.

Parameters

• affine (torch Tensor) – Nxdxd batched square matrix

• src_size (sequence of int) – source image spatial shape

• dst_size (sequence of int) – target image spatial shape

• align_corners (bool) – if True, consider -1 and 1 to refer to the centers of the corner pixels rather than the image corners. See also: https://pytorch.org/docs/stable/nn.functional.html#torch.nn.functional.grid_sample

Raises

• ValueError – affine must be a tensor

• ValueError – affine must be Nxdxd, got {tuple(affine.shape)}

• ValueError – affine suggests a {sr}-D transform, but the sizes are src_size={src_size}, dst_size={dst_size}