Network architectures¶
Blocks¶
ADN¶
- class monai.networks.blocks.ADN(ordering='NDA', in_channels=None, act='RELU', norm=None, norm_dim=None, dropout=None, dropout_dim=None)[source]¶
Constructs a sequential module of optional activation, dropout, and normalization layers (with an arbitrary order):
-- (Norm) -- (Dropout) -- (Acti) --
- Parameters
ordering (
str
) – a string representing the ordering of activation, dropout, and normalization. Defaults to “NDA”.in_channels (
Optional
[int
]) – C from an expected input of size (N, C, H[, W, D]).act (
Union
[Tuple
,str
,None
]) – activation type and arguments. Defaults to PReLU.norm (
Union
[Tuple
,str
,None
]) – feature normalization type and arguments. Defaults to instance norm.norm_dim (
Optional
[int
]) – determine the spatial dimensions of the normalization layer. defaults to dropout_dim if unspecified.dropout (
Union
[Tuple
,str
,float
,None
]) – dropout ratio. Defaults to no dropout.dropout_dim (
Optional
[int
]) –determine the spatial dimensions of dropout. defaults to norm_dim if unspecified.
When dropout_dim = 1, randomly zeroes some of the elements for each channel.
When dropout_dim = 2, Randomly zeroes out entire channels (a channel is a 2D feature map).
When dropout_dim = 3, Randomly zeroes out entire channels (a channel is a 3D feature map).
Examples:
# activation, group norm, dropout >>> norm_params = ("GROUP", {"num_groups": 1, "affine": False}) >>> ADN(norm=norm_params, in_channels=1, dropout_dim=1, dropout=0.8, ordering="AND") ADN( (A): ReLU() (N): GroupNorm(1, 1, eps=1e-05, affine=False) (D): Dropout(p=0.8, inplace=False) ) # LeakyReLU, dropout >>> act_params = ("leakyrelu", {"negative_slope": 0.1, "inplace": True}) >>> ADN(act=act_params, in_channels=1, dropout_dim=1, dropout=0.8, ordering="AD") ADN( (A): LeakyReLU(negative_slope=0.1, inplace=True) (D): Dropout(p=0.8, inplace=False) )
See also
monai.networks.layers.Dropout
monai.networks.layers.Act
monai.networks.layers.Norm
monai.networks.layers.split_args
Initializes internal Module state, shared by both nn.Module and ScriptModule.
Convolution¶
- class monai.networks.blocks.Convolution(spatial_dims, in_channels, out_channels, strides=1, kernel_size=3, adn_ordering='NDA', act='PRELU', norm='INSTANCE', dropout=None, dropout_dim=1, dilation=1, groups=1, bias=True, conv_only=False, is_transposed=False, padding=None, output_padding=None, dimensions=None)[source]¶
Constructs a convolution with normalization, optional dropout, and optional activation layers:
-- (Conv|ConvTrans) -- (Norm -- Dropout -- Acti) --
if
conv_only
set toTrue
:-- (Conv|ConvTrans) --
For example:
from monai.networks.blocks import Convolution conv = Convolution( dimensions=3, in_channels=1, out_channels=1, adn_ordering="ADN", act=("prelu", {"init": 0.2}), dropout=0.1, norm=("layer", {"normalized_shape": (10, 10, 10)}), ) print(conv)
output:
Convolution( (conv): Conv3d(1, 1, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1)) (adn): ADN( (A): PReLU(num_parameters=1) (D): Dropout(p=0.1, inplace=False) (N): LayerNorm((10, 10, 10), eps=1e-05, elementwise_affine=True) ) )
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.strides (
Union
[Sequence
[int
],int
]) – convolution stride. Defaults to 1.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size. Defaults to 3.adn_ordering (
str
) – a string representing the ordering of activation, normalization, and dropout. Defaults to “NDA”.act (
Union
[Tuple
,str
,None
]) – activation type and arguments. Defaults to PReLU.norm (
Union
[Tuple
,str
,None
]) – feature normalization type and arguments. Defaults to instance norm.dropout (
Union
[Tuple
,str
,float
,None
]) – dropout ratio. Defaults to no dropout.dropout_dim (
Optional
[int
]) –determine the spatial dimensions of dropout. Defaults to 1.
When dropout_dim = 1, randomly zeroes some of the elements for each channel.
When dropout_dim = 2, Randomly zeroes out entire channels (a channel is a 2D feature map).
When dropout_dim = 3, Randomly zeroes out entire channels (a channel is a 3D feature map).
The value of dropout_dim should be no no larger than the value of spatial_dims.
dilation (
Union
[Sequence
[int
],int
]) – dilation rate. Defaults to 1.groups (
int
) – controls the connections between inputs and outputs. Defaults to 1.bias (
bool
) – whether to have a bias term. Defaults to True.conv_only (
bool
) – whether to use the convolutional layer only. Defaults to False.is_transposed (
bool
) – if True uses ConvTrans instead of Conv. Defaults to False.padding (
Union
[Sequence
[int
],int
,None
]) – controls the amount of implicit zero-paddings on both sides for padding number of points for each dimension. Defaults to None.output_padding (
Union
[Sequence
[int
],int
,None
]) – controls the additional size added to one side of the output shape. Defaults to None.
Deprecated since version 0.6.0:
dimensions
is deprecated, usespatial_dims
instead.Initializes internal Module state, shared by both nn.Module and ScriptModule.
CRF¶
- class monai.networks.blocks.CRF(iterations=5, bilateral_weight=1.0, gaussian_weight=1.0, bilateral_spatial_sigma=5.0, bilateral_color_sigma=0.5, gaussian_spatial_sigma=5.0, update_factor=3.0, compatibility_matrix=None)[source]¶
Conditional Random Field: Combines message passing with a class compatibility convolution into an iterative process designed to successively minimise the energy of the class labeling.
In this implementation, the message passing step is a weighted combination of a gaussian filter and a bilateral filter. The bilateral term is included to respect existing structure within the reference tensor.
- Parameters
iterations (
int
) – the number of iterations.bilateral_weight (
float
) – the weighting of the bilateral term in the message passing step.gaussian_weight (
float
) – the weighting of the gaussian term in the message passing step.bilateral_spatial_sigma (
float
) – standard deviation in spatial coordinates for the bilateral term.bilateral_color_sigma (
float
) – standard deviation in color space for the bilateral term.gaussian_spatial_sigma (
float
) – standard deviation in spatial coordinates for the gaussian term.update_factor (
float
) – determines the magnitude of each update.compatibility_matrix (
Optional
[Tensor
]) – a matrix describing class compatibility, should be NxN where N is the number of classes.
- __init__(iterations=5, bilateral_weight=1.0, gaussian_weight=1.0, bilateral_spatial_sigma=5.0, bilateral_color_sigma=0.5, gaussian_spatial_sigma=5.0, update_factor=3.0, compatibility_matrix=None)[source]¶
- Parameters
iterations (
int
) – the number of iterations.bilateral_weight (
float
) – the weighting of the bilateral term in the message passing step.gaussian_weight (
float
) – the weighting of the gaussian term in the message passing step.bilateral_spatial_sigma (
float
) – standard deviation in spatial coordinates for the bilateral term.bilateral_color_sigma (
float
) – standard deviation in color space for the bilateral term.gaussian_spatial_sigma (
float
) – standard deviation in spatial coordinates for the gaussian term.update_factor (
float
) – determines the magnitude of each update.compatibility_matrix (
Optional
[Tensor
]) – a matrix describing class compatibility, should be NxN where N is the number of classes.
ResidualUnit¶
- class monai.networks.blocks.ResidualUnit(spatial_dims, in_channels, out_channels, strides=1, kernel_size=3, subunits=2, adn_ordering='NDA', act='PRELU', norm='INSTANCE', dropout=None, dropout_dim=1, dilation=1, bias=True, last_conv_only=False, padding=None, dimensions=None)[source]¶
Residual module with multiple convolutions and a residual connection.
For example:
from monai.networks.blocks import ResidualUnit convs = ResidualUnit( spatial_dims=3, in_channels=1, out_channels=1, adn_ordering="AN", act=("prelu", {"init": 0.2}), norm=("layer", {"normalized_shape": (10, 10, 10)}), ) print(convs)
output:
ResidualUnit( (conv): Sequential( (unit0): Convolution( (conv): Conv3d(1, 1, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1)) (adn): ADN( (A): PReLU(num_parameters=1) (N): LayerNorm((10, 10, 10), eps=1e-05, elementwise_affine=True) ) ) (unit1): Convolution( (conv): Conv3d(1, 1, kernel_size=(3, 3, 3), stride=(1, 1, 1), padding=(1, 1, 1)) (adn): ADN( (A): PReLU(num_parameters=1) (N): LayerNorm((10, 10, 10), eps=1e-05, elementwise_affine=True) ) ) ) (residual): Identity() )
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.strides (
Union
[Sequence
[int
],int
]) – convolution stride. Defaults to 1.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size. Defaults to 3.subunits (
int
) – number of convolutions. Defaults to 2.adn_ordering (
str
) – a string representing the ordering of activation, normalization, and dropout. Defaults to “NDA”.act (
Union
[Tuple
,str
,None
]) – activation type and arguments. Defaults to PReLU.norm (
Union
[Tuple
,str
,None
]) – feature normalization type and arguments. Defaults to instance norm.dropout (
Union
[Tuple
,str
,float
,None
]) – dropout ratio. Defaults to no dropout.dropout_dim (
Optional
[int
]) –determine the dimensions of dropout. Defaults to 1.
When dropout_dim = 1, randomly zeroes some of the elements for each channel.
When dropout_dim = 2, Randomly zero out entire channels (a channel is a 2D feature map).
When dropout_dim = 3, Randomly zero out entire channels (a channel is a 3D feature map).
The value of dropout_dim should be no no larger than the value of dimensions.
dilation (
Union
[Sequence
[int
],int
]) – dilation rate. Defaults to 1.bias (
bool
) – whether to have a bias term. Defaults to True.last_conv_only (
bool
) – for the last subunit, whether to use the convolutional layer only. Defaults to False.padding (
Union
[Sequence
[int
],int
,None
]) – controls the amount of implicit zero-paddings on both sides for padding number of points for each dimension. Defaults to None.
Deprecated since version 0.6.0:
dimensions
is deprecated, usespatial_dims
instead.See also
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Tensor
Swish¶
- class monai.networks.blocks.Swish(alpha=1.0)[source]¶
Applies the element-wise function:
\[\text{Swish}(x) = x * \text{Sigmoid}(\alpha * x) ~~~~\text{for constant value}~ \alpha.\]Citation: Searching for Activation Functions, Ramachandran et al., 2017, https://arxiv.org/abs/1710.05941.
- Shape:
Input: \((N, *)\) where * means, any number of additional dimensions
Output: \((N, *)\), same shape as the input
Examples:
>>> import torch >>> from monai.networks.layers.factories import Act >>> m = Act['swish']() >>> input = torch.randn(2) >>> output = m(input)
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(input)[source]¶
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.- Return type
Tensor
MemoryEfficientSwish¶
- class monai.networks.blocks.MemoryEfficientSwish(inplace=False)[source]¶
Applies the element-wise function:
\[\text{Swish}(x) = x * \text{Sigmoid}(\alpha * x) ~~~~\text{for constant value}~ \alpha=1.\]Memory efficient implementation for training following recommendation from: https://github.com/lukemelas/EfficientNet-PyTorch/issues/18#issuecomment-511677853
Results in ~ 30% memory saving during training as compared to Swish()
Citation: Searching for Activation Functions, Ramachandran et al., 2017, https://arxiv.org/abs/1710.05941.
From Pytorch 1.7.0+, the optimized version of Swish named SiLU is implemented, this class will utilize torch.nn.functional.silu to do the calculation if meets the version.
- Shape:
Input: \((N, *)\) where * means, any number of additional dimensions
Output: \((N, *)\), same shape as the input
Examples:
>>> import torch >>> from monai.networks.layers.factories import Act >>> m = Act['memswish']() >>> input = torch.randn(2) >>> output = m(input)
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(input)[source]¶
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.
Mish¶
- class monai.networks.blocks.Mish(inplace=False)[source]¶
Applies the element-wise function:
\[\text{Mish}(x) = x * tanh(\text{softplus}(x)).\]Citation: Mish: A Self Regularized Non-Monotonic Activation Function, Diganta Misra, 2019, https://arxiv.org/abs/1908.08681.
From Pytorch 1.9.0+, the optimized version of Mish is implemented, this class will utilize torch.nn.functional.mish to do the calculation if meets the version.
- Shape:
Input: \((N, *)\) where * means, any number of additional dimensions
Output: \((N, *)\), same shape as the input
Examples:
>>> import torch >>> from monai.networks.layers.factories import Act >>> m = Act['mish']() >>> input = torch.randn(2) >>> output = m(input)
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(input)[source]¶
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.
GCN Module¶
- class monai.networks.blocks.GCN(inplanes, planes, ks=7)[source]¶
The Global Convolutional Network module using large 1D Kx1 and 1xK kernels to represent 2D kernels.
- Parameters
inplanes (
int
) – number of input channels.planes (
int
) – number of output channels.ks (
int
) – kernel size for one dimension. Defaults to 7.
Refinement Module¶
FCN Module¶
- class monai.networks.blocks.FCN(out_channels=1, upsample_mode='bilinear', pretrained=True, progress=True)[source]¶
2D FCN network with 3 input channels. The small decoder is built with the GCN and Refine modules. The code is adapted from lsqshr’s official 2D code.
- Parameters
out_channels (
int
) – number of output channels. Defaults to 1.upsample_mode (
str
) –[
"transpose"
,"bilinear"
] The mode of upsampling manipulations. Using the second mode cannot guarantee the model’s reproducibility. Defaults tobilinear
.transpose
, uses transposed convolution layers.bilinear
, uses bilinear interpolation.
pretrained (
bool
) – If True, returns a model pre-trained on ImageNetprogress (
bool
) – If True, displays a progress bar of the download to stderr.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
Multi-Channel FCN Module¶
- class monai.networks.blocks.MCFCN(in_channels=3, out_channels=1, upsample_mode='bilinear', pretrained=True, progress=True)[source]¶
The multi-channel version of the 2D FCN module. Adds a projection layer to take arbitrary number of inputs.
- Parameters
in_channels (
int
) – number of input channels. Defaults to 3.out_channels (
int
) – number of output channels. Defaults to 1.upsample_mode (
str
) –[
"transpose"
,"bilinear"
] The mode of upsampling manipulations. Using the second mode cannot guarantee the model’s reproducibility. Defaults tobilinear
.transpose
, uses transposed convolution layers.bilinear
, uses bilinear interpolate.
pretrained (
bool
) – If True, returns a model pre-trained on ImageNetprogress (
bool
) – If True, displays a progress bar of the download to stderr.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
Dynamic-Unet Block¶
- class monai.networks.blocks.UnetResBlock(spatial_dims, in_channels, out_channels, kernel_size, stride, norm_name, act_name=('leakyrelu', {'inplace': True, 'negative_slope': 0.01}), dropout=None)[source]¶
A skip-connection based module that can be used for DynUNet, based on: Automated Design of Deep Learning Methods for Biomedical Image Segmentation. nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation.
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size.stride (
Union
[Sequence
[int
],int
]) – convolution stride.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.act_name (
Union
[Tuple
,str
]) – activation layer type and arguments.dropout (
Union
[Tuple
,str
,float
,None
]) – dropout probability.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(inp)[source]¶
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.blocks.UnetBasicBlock(spatial_dims, in_channels, out_channels, kernel_size, stride, norm_name, act_name=('leakyrelu', {'inplace': True, 'negative_slope': 0.01}), dropout=None)[source]¶
A CNN module module that can be used for DynUNet, based on: Automated Design of Deep Learning Methods for Biomedical Image Segmentation. nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation.
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size.stride (
Union
[Sequence
[int
],int
]) – convolution stride.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.act_name (
Union
[Tuple
,str
]) – activation layer type and arguments.dropout (
Union
[Tuple
,str
,float
,None
]) – dropout probability.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(inp)[source]¶
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.blocks.UnetUpBlock(spatial_dims, in_channels, out_channels, kernel_size, stride, upsample_kernel_size, norm_name, act_name=('leakyrelu', {'inplace': True, 'negative_slope': 0.01}), dropout=None, trans_bias=False)[source]¶
An upsampling module that can be used for DynUNet, based on: Automated Design of Deep Learning Methods for Biomedical Image Segmentation. nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation.
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size.stride (
Union
[Sequence
[int
],int
]) – convolution stride.upsample_kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size for transposed convolution layers.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.act_name (
Union
[Tuple
,str
]) – activation layer type and arguments.dropout (
Union
[Tuple
,str
,float
,None
]) – dropout probability.trans_bias (
bool
) – transposed convolution bias.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(inp, skip)[source]¶
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.blocks.UnetOutBlock(spatial_dims, in_channels, out_channels, dropout=None)[source]¶
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(inp)[source]¶
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.
SegResnet Block¶
- class monai.networks.blocks.ResBlock(spatial_dims, in_channels, norm, kernel_size=3, act=('RELU', {'inplace': True}))[source]¶
ResBlock employs skip connection and two convolution blocks and is used in SegResNet based on 3D MRI brain tumor segmentation using autoencoder regularization.
- Parameters
spatial_dims (
int
) – number of spatial dimensions, could be 1, 2 or 3.in_channels (
int
) – number of input channels.norm (
Union
[Tuple
,str
]) – feature normalization type and arguments.kernel_size (
int
) – convolution kernel size, the value should be an odd number. Defaults to 3.act (
Union
[Tuple
,str
]) – activation type and arguments. Defaults toRELU
.
- __init__(spatial_dims, in_channels, norm, kernel_size=3, act=('RELU', {'inplace': True}))[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions, could be 1, 2 or 3.in_channels (
int
) – number of input channels.norm (
Union
[Tuple
,str
]) – feature normalization type and arguments.kernel_size (
int
) – convolution kernel size, the value should be an odd number. Defaults to 3.act (
Union
[Tuple
,str
]) – activation type and arguments. Defaults toRELU
.
- forward(x)[source]¶
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.
SABlock Block¶
- class monai.networks.blocks.SABlock(hidden_size, num_heads, dropout_rate=0.0)[source]¶
A self-attention block, based on: “Dosovitskiy et al., An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale <https://arxiv.org/abs/2010.11929>”
- Parameters
hidden_size (
int
) – dimension of hidden layer.num_heads (
int
) – number of attention heads.dropout_rate (
float
) – faction of the input units to drop.
- __init__(hidden_size, num_heads, dropout_rate=0.0)[source]¶
- Parameters
hidden_size (
int
) – dimension of hidden layer.num_heads (
int
) – number of attention heads.dropout_rate (
float
) – faction of the input units to drop.
- forward(x)[source]¶
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', {'inplace': True}), acti_type_2='sigmoid', add_residual=False)[source]¶
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 (
Union
[Tuple
[str
,Dict
],str
]) – activation type of the hidden squeeze layer. Defaults to("relu", {"inplace": True})
.acti_type_2 (
Union
[Tuple
[str
,Dict
],str
]) – activation type of the output squeeze layer. Defaults to “sigmoid”.
- Raises
ValueError – When
r
is nonpositive or larger thanin_channels
.
See also
- __init__(spatial_dims, in_channels, r=2, acti_type_1=('relu', {'inplace': True}), acti_type_2='sigmoid', add_residual=False)[source]¶
- 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 (
Union
[Tuple
[str
,Dict
],str
]) – activation type of the hidden squeeze layer. Defaults to("relu", {"inplace": True})
.acti_type_2 (
Union
[Tuple
[str
,Dict
],str
]) – activation type of the output squeeze layer. Defaults to “sigmoid”.
- Raises
ValueError – When
r
is nonpositive or larger thanin_channels
.
See also
Transformer Block¶
- class monai.networks.blocks.TransformerBlock(hidden_size, mlp_dim, num_heads, dropout_rate=0.0)[source]¶
A transformer block, based on: “Dosovitskiy et al., An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale <https://arxiv.org/abs/2010.11929>”
- Parameters
hidden_size (
int
) – dimension of hidden layer.mlp_dim (
int
) – dimension of feedforward layer.num_heads (
int
) – number of attention heads.dropout_rate (
float
) – faction of the input units to drop.
- __init__(hidden_size, mlp_dim, num_heads, dropout_rate=0.0)[source]¶
- Parameters
hidden_size (
int
) – dimension of hidden layer.mlp_dim (
int
) – dimension of feedforward layer.num_heads (
int
) – number of attention heads.dropout_rate (
float
) – faction of the input units to drop.
- forward(x)[source]¶
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.
UNETR Block¶
- class monai.networks.blocks.UnetrBasicBlock(spatial_dims, in_channels, out_channels, kernel_size, stride, norm_name, res_block=False)[source]¶
A CNN module that can be used for UNETR, based on: “Hatamizadeh et al., UNETR: Transformers for 3D Medical Image Segmentation <https://arxiv.org/abs/2103.10504>”
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size.stride (
Union
[Sequence
[int
],int
]) – convolution stride.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.res_block (
bool
) – bool argument to determine if residual block is used.
- __init__(spatial_dims, in_channels, out_channels, kernel_size, stride, norm_name, res_block=False)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size.stride (
Union
[Sequence
[int
],int
]) – convolution stride.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.res_block (
bool
) – bool argument to determine if residual block is used.
- forward(inp)[source]¶
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.blocks.UnetrUpBlock(spatial_dims, in_channels, out_channels, kernel_size, upsample_kernel_size, norm_name, res_block=False)[source]¶
An upsampling module that can be used for UNETR: “Hatamizadeh et al., UNETR: Transformers for 3D Medical Image Segmentation <https://arxiv.org/abs/2103.10504>”
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size.upsample_kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size for transposed convolution layers.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.res_block (
bool
) – bool argument to determine if residual block is used.
- __init__(spatial_dims, in_channels, out_channels, kernel_size, upsample_kernel_size, norm_name, res_block=False)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size.upsample_kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size for transposed convolution layers.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.res_block (
bool
) – bool argument to determine if residual block is used.
- forward(inp, skip)[source]¶
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.blocks.UnetrPrUpBlock(spatial_dims, in_channels, out_channels, num_layer, kernel_size, stride, upsample_kernel_size, norm_name, conv_block=False, res_block=False)[source]¶
A projection upsampling module that can be used for UNETR: “Hatamizadeh et al., UNETR: Transformers for 3D Medical Image Segmentation <https://arxiv.org/abs/2103.10504>”
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.num_layer (
int
) – number of upsampling blocks.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size.stride (
Union
[Sequence
[int
],int
]) – convolution stride.upsample_kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size for transposed convolution layers.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.conv_block (
bool
) – bool argument to determine if convolutional block is used.res_block (
bool
) – bool argument to determine if residual block is used.
- __init__(spatial_dims, in_channels, out_channels, num_layer, kernel_size, stride, upsample_kernel_size, norm_name, conv_block=False, res_block=False)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.num_layer (
int
) – number of upsampling blocks.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size.stride (
Union
[Sequence
[int
],int
]) – convolution stride.upsample_kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size for transposed convolution layers.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.conv_block (
bool
) – bool argument to determine if convolutional block is used.res_block (
bool
) – bool argument to determine if residual block is used.
- forward(x)[source]¶
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.
Residual Squeeze-and-Excitation¶
- class monai.networks.blocks.ResidualSELayer(spatial_dims, in_channels, r=2, acti_type_1='leakyrelu', acti_type_2='relu')[source]¶
A “squeeze-and-excitation”-like layer with a residual connection:
--+-- SE --o-- | | +--------+
- 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 (
Union
[Tuple
[str
,Dict
],str
]) – defaults to “leakyrelu”.acti_type_2 (
Union
[Tuple
[str
,Dict
],str
]) – defaults to “relu”.
See also
- __init__(spatial_dims, in_channels, r=2, acti_type_1='leakyrelu', acti_type_2='relu')[source]¶
- 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 (
Union
[Tuple
[str
,Dict
],str
]) – defaults to “leakyrelu”.acti_type_2 (
Union
[Tuple
[str
,Dict
],str
]) – defaults to “relu”.
See also
Squeeze-and-Excitation Block¶
- class monai.networks.blocks.SEBlock(spatial_dims, in_channels, n_chns_1, n_chns_2, n_chns_3, conv_param_1=None, conv_param_2=None, conv_param_3=None, project=None, r=2, acti_type_1=('relu', {'inplace': True}), acti_type_2='sigmoid', acti_type_final=('relu', {'inplace': True}))[source]¶
Residual module enhanced with Squeeze-and-Excitation:
----+- conv1 -- conv2 -- conv3 -- SE -o--- | | +---(channel project if needed)----+
Re-implementation of the SE-Resnet 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.n_chns_1 (
int
) – number of output channels in the 1st convolution.n_chns_2 (
int
) – number of output channels in the 2nd convolution.n_chns_3 (
int
) – number of output channels in the 3rd convolution.conv_param_1 (
Optional
[Dict
]) – additional parameters to the 1st convolution. Defaults to{"kernel_size": 1, "norm": Norm.BATCH, "act": ("relu", {"inplace": True})}
conv_param_2 (
Optional
[Dict
]) – additional parameters to the 2nd convolution. Defaults to{"kernel_size": 3, "norm": Norm.BATCH, "act": ("relu", {"inplace": True})}
conv_param_3 (
Optional
[Dict
]) – additional parameters to the 3rd convolution. Defaults to{"kernel_size": 1, "norm": Norm.BATCH, "act": None}
project (
Optional
[Convolution
]) – in the case of residual chns and output chns doesn’t match, a project (Conv) layer/block is used to adjust the number of chns. In SENET, it is consisted with a Conv layer as well as a Norm layer. Defaults to None (chns are matchable) or a Conv layer with kernel size 1.r (
int
) – the reduction ratio r in the paper. Defaults to 2.acti_type_1 (
Union
[Tuple
[str
,Dict
],str
]) – activation type of the hidden squeeze layer. Defaults to “relu”.acti_type_2 (
Union
[Tuple
[str
,Dict
],str
]) – activation type of the output squeeze layer. Defaults to “sigmoid”.acti_type_final (
Union
[Tuple
[str
,Dict
],str
,None
]) – activation type of the end of the block. Defaults to “relu”.
See also
- __init__(spatial_dims, in_channels, n_chns_1, n_chns_2, n_chns_3, conv_param_1=None, conv_param_2=None, conv_param_3=None, project=None, r=2, acti_type_1=('relu', {'inplace': True}), acti_type_2='sigmoid', acti_type_final=('relu', {'inplace': True}))[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions, could be 1, 2, or 3.in_channels (
int
) – number of input channels.n_chns_1 (
int
) – number of output channels in the 1st convolution.n_chns_2 (
int
) – number of output channels in the 2nd convolution.n_chns_3 (
int
) – number of output channels in the 3rd convolution.conv_param_1 (
Optional
[Dict
]) – additional parameters to the 1st convolution. Defaults to{"kernel_size": 1, "norm": Norm.BATCH, "act": ("relu", {"inplace": True})}
conv_param_2 (
Optional
[Dict
]) – additional parameters to the 2nd convolution. Defaults to{"kernel_size": 3, "norm": Norm.BATCH, "act": ("relu", {"inplace": True})}
conv_param_3 (
Optional
[Dict
]) – additional parameters to the 3rd convolution. Defaults to{"kernel_size": 1, "norm": Norm.BATCH, "act": None}
project (
Optional
[Convolution
]) – in the case of residual chns and output chns doesn’t match, a project (Conv) layer/block is used to adjust the number of chns. In SENET, it is consisted with a Conv layer as well as a Norm layer. Defaults to None (chns are matchable) or a Conv layer with kernel size 1.r (
int
) – the reduction ratio r in the paper. Defaults to 2.acti_type_1 (
Union
[Tuple
[str
,Dict
],str
]) – activation type of the hidden squeeze layer. Defaults to “relu”.acti_type_2 (
Union
[Tuple
[str
,Dict
],str
]) – activation type of the output squeeze layer. Defaults to “sigmoid”.acti_type_final (
Union
[Tuple
[str
,Dict
],str
,None
]) – activation type of the end of the block. Defaults to “relu”.
See also
Squeeze-and-Excitation Bottleneck¶
- class monai.networks.blocks.SEBottleneck(spatial_dims, inplanes, planes, groups, reduction, stride=1, downsample=None)[source]¶
Bottleneck for SENet154.
- Parameters
spatial_dims (
int
) – number of spatial dimensions, could be 1, 2, or 3.in_channels – number of input channels.
n_chns_1 – number of output channels in the 1st convolution.
n_chns_2 – number of output channels in the 2nd convolution.
n_chns_3 – number of output channels in the 3rd convolution.
conv_param_1 – additional parameters to the 1st convolution. Defaults to
{"kernel_size": 1, "norm": Norm.BATCH, "act": ("relu", {"inplace": True})}
conv_param_2 – additional parameters to the 2nd convolution. Defaults to
{"kernel_size": 3, "norm": Norm.BATCH, "act": ("relu", {"inplace": True})}
conv_param_3 – additional parameters to the 3rd convolution. Defaults to
{"kernel_size": 1, "norm": Norm.BATCH, "act": None}
project – in the case of residual chns and output chns doesn’t match, a project (Conv) layer/block is used to adjust the number of chns. In SENET, it is consisted with a Conv layer as well as a Norm layer. Defaults to None (chns are matchable) or a Conv layer with kernel size 1.
r – the reduction ratio r in the paper. Defaults to 2.
acti_type_1 – activation type of the hidden squeeze layer. Defaults to “relu”.
acti_type_2 – activation type of the output squeeze layer. Defaults to “sigmoid”.
acti_type_final – activation type of the end of the block. Defaults to “relu”.
See also
Squeeze-and-Excitation Resnet Bottleneck¶
- class monai.networks.blocks.SEResNetBottleneck(spatial_dims, inplanes, planes, groups, reduction, stride=1, downsample=None)[source]¶
ResNet bottleneck with a Squeeze-and-Excitation module. It follows Caffe implementation and uses strides=stride in conv1 and not in conv2 (the latter is used in the torchvision implementation of ResNet).
- Parameters
spatial_dims (
int
) – number of spatial dimensions, could be 1, 2, or 3.in_channels – number of input channels.
n_chns_1 – number of output channels in the 1st convolution.
n_chns_2 – number of output channels in the 2nd convolution.
n_chns_3 – number of output channels in the 3rd convolution.
conv_param_1 – additional parameters to the 1st convolution. Defaults to
{"kernel_size": 1, "norm": Norm.BATCH, "act": ("relu", {"inplace": True})}
conv_param_2 – additional parameters to the 2nd convolution. Defaults to
{"kernel_size": 3, "norm": Norm.BATCH, "act": ("relu", {"inplace": True})}
conv_param_3 – additional parameters to the 3rd convolution. Defaults to
{"kernel_size": 1, "norm": Norm.BATCH, "act": None}
project – in the case of residual chns and output chns doesn’t match, a project (Conv) layer/block is used to adjust the number of chns. In SENET, it is consisted with a Conv layer as well as a Norm layer. Defaults to None (chns are matchable) or a Conv layer with kernel size 1.
r – the reduction ratio r in the paper. Defaults to 2.
acti_type_1 – activation type of the hidden squeeze layer. Defaults to “relu”.
acti_type_2 – activation type of the output squeeze layer. Defaults to “sigmoid”.
acti_type_final – activation type of the end of the block. Defaults to “relu”.
See also
Squeeze-and-Excitation ResNeXt Bottleneck¶
- class monai.networks.blocks.SEResNeXtBottleneck(spatial_dims, inplanes, planes, groups, reduction, stride=1, downsample=None, base_width=4)[source]¶
ResNeXt bottleneck type C with a Squeeze-and-Excitation module.
- Parameters
spatial_dims (
int
) – number of spatial dimensions, could be 1, 2, or 3.in_channels – number of input channels.
n_chns_1 – number of output channels in the 1st convolution.
n_chns_2 – number of output channels in the 2nd convolution.
n_chns_3 – number of output channels in the 3rd convolution.
conv_param_1 – additional parameters to the 1st convolution. Defaults to
{"kernel_size": 1, "norm": Norm.BATCH, "act": ("relu", {"inplace": True})}
conv_param_2 – additional parameters to the 2nd convolution. Defaults to
{"kernel_size": 3, "norm": Norm.BATCH, "act": ("relu", {"inplace": True})}
conv_param_3 – additional parameters to the 3rd convolution. Defaults to
{"kernel_size": 1, "norm": Norm.BATCH, "act": None}
project – in the case of residual chns and output chns doesn’t match, a project (Conv) layer/block is used to adjust the number of chns. In SENET, it is consisted with a Conv layer as well as a Norm layer. Defaults to None (chns are matchable) or a Conv layer with kernel size 1.
r – the reduction ratio r in the paper. Defaults to 2.
acti_type_1 – activation type of the hidden squeeze layer. Defaults to “relu”.
acti_type_2 – activation type of the output squeeze layer. Defaults to “sigmoid”.
acti_type_final – activation type of the end of the block. Defaults to “relu”.
See also
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', bias=False)[source]¶
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 (
Sequence
[int
]) – a sequence of four convolutional kernel sizes. Defaults to (1, 3, 3, 3) for four (dilated) convolutions.dilations (
Sequence
[int
]) – a sequence of four convolutional dilation parameters. Defaults to (1, 2, 4, 6) for four (dilated) convolutions.norm_type (
Union
[Tuple
,str
,None
]) – final kernel-size-one convolution normalization type. Defaults to batch norm.acti_type (
Union
[Tuple
,str
,None
]) – final kernel-size-one convolution activation type. Defaults to leaky ReLU.bias (
bool
) – whether to have a bias term in convolution blocks. Defaults to False. According to Performance Tuning Guide, if a conv layer is directly followed by a batch norm layer, bias should be False.
- Raises
ValueError – When
kernel_sizes
length differs fromdilations
.
- __init__(spatial_dims, in_channels, conv_out_channels, kernel_sizes=(1, 3, 3, 3), dilations=(1, 2, 4, 6), norm_type='BATCH', acti_type='LEAKYRELU', bias=False)[source]¶
- 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 (
Sequence
[int
]) – a sequence of four convolutional kernel sizes. Defaults to (1, 3, 3, 3) for four (dilated) convolutions.dilations (
Sequence
[int
]) – a sequence of four convolutional dilation parameters. Defaults to (1, 2, 4, 6) for four (dilated) convolutions.norm_type (
Union
[Tuple
,str
,None
]) – final kernel-size-one convolution normalization type. Defaults to batch norm.acti_type (
Union
[Tuple
,str
,None
]) – final kernel-size-one convolution activation type. Defaults to leaky ReLU.bias (
bool
) –whether to have a bias term in convolution blocks. Defaults to False. According to Performance Tuning Guide, if a conv layer is directly followed by a batch norm layer, bias should be False.
- Raises
ValueError – When
kernel_sizes
length differs fromdilations
.
MaxAvgPooling¶
- class monai.networks.blocks.MaxAvgPool(spatial_dims, kernel_size, stride=None, padding=0, ceil_mode=False)[source]¶
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 (
Union
[Sequence
[int
],int
]) – the kernel size of both pooling operations.stride (
Union
[Sequence
[int
],int
,None
]) – the stride of the window. Default value is kernel_size.padding (
Union
[Sequence
[int
],int
]) – 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.
- __init__(spatial_dims, kernel_size, stride=None, padding=0, ceil_mode=False)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions of the input image.kernel_size (
Union
[Sequence
[int
],int
]) – the kernel size of both pooling operations.stride (
Union
[Sequence
[int
],int
,None
]) – the stride of the window. Default value is kernel_size.padding (
Union
[Sequence
[int
],int
]) – 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.
Upsampling¶
- class monai.networks.blocks.UpSample(spatial_dims, in_channels=None, out_channels=None, scale_factor=2, size=None, mode=UpsampleMode.DECONV, pre_conv='default', interp_mode=InterpolateMode.LINEAR, align_corners=True, bias=True, apply_pad_pool=True, dimensions=None)[source]¶
Upsamples data by scale_factor. Supported modes are:
“deconv”: uses a transposed convolution.
“nontrainable”: uses
torch.nn.Upsample
.“pixelshuffle”: uses
monai.networks.blocks.SubpixelUpsample
.
This module can optionally take a pre-convolution (often used to map the number of features from in_channels to out_channels).
- Parameters
spatial_dims (
int
) – number of spatial dimensions of the input image.in_channels (
Optional
[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 (
Union
[Sequence
[float
],float
]) – multiplier for spatial size. Has to match input size if it is a tuple. Defaults to 2.size (
Union
[Tuple
[int
],int
,None
]) – spatial size of the output image. Only used whenmode
isUpsampleMode.NONTRAINABLE
. In torch.nn.functional.interpolate, only one of size or scale_factor should be defined, thus if size is defined, scale_factor will not be used. Defaults to None.mode (
Union
[UpsampleMode
,str
]) – {"deconv"
,"nontrainable"
,"pixelshuffle"
}. Defaults to"deconv"
.pre_conv (
Union
[Module
,str
,None
]) – a conv block applied before upsampling. Defaults to “default”. Whenconv_block
is"default"
, one reserved conv layer will be utilized when Only used in the “nontrainable” or “pixelshuffle” mode.interp_mode (
Union
[InterpolateMode
,str
]) – {"nearest"
,"linear"
,"bilinear"
,"bicubic"
,"trilinear"
} Only used in the “nontrainable” mode. If ends with"linear"
will usespatial 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/generated/torch.nn.Upsample.htmlalign_corners (
Optional
[bool
]) – set the align_corners parameter of torch.nn.Upsample. Defaults to True. Only used in the “nontrainable” mode.bias (
bool
) – whether to have a bias term in the default preconv and deconv layers. Defaults to True.apply_pad_pool (
bool
) – if True the upsampled tensor is padded then average pooling is applied with a kernel the size of scale_factor with a stride of 1. See also:monai.networks.blocks.SubpixelUpsample
. Only used in the “pixelshuffle” mode.
Deprecated since version 0.6.0:
dimensions
is deprecated, usespatial_dims
instead.- __init__(spatial_dims, in_channels=None, out_channels=None, scale_factor=2, size=None, mode=UpsampleMode.DECONV, pre_conv='default', interp_mode=InterpolateMode.LINEAR, align_corners=True, bias=True, apply_pad_pool=True, dimensions=None)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions of the input image.in_channels (
Optional
[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 (
Union
[Sequence
[float
],float
]) – multiplier for spatial size. Has to match input size if it is a tuple. Defaults to 2.size (
Union
[Tuple
[int
],int
,None
]) – spatial size of the output image. Only used whenmode
isUpsampleMode.NONTRAINABLE
. In torch.nn.functional.interpolate, only one of size or scale_factor should be defined, thus if size is defined, scale_factor will not be used. Defaults to None.mode (
Union
[UpsampleMode
,str
]) – {"deconv"
,"nontrainable"
,"pixelshuffle"
}. Defaults to"deconv"
.pre_conv (
Union
[Module
,str
,None
]) – a conv block applied before upsampling. Defaults to “default”. Whenconv_block
is"default"
, one reserved conv layer will be utilized when Only used in the “nontrainable” or “pixelshuffle” mode.interp_mode (
Union
[InterpolateMode
,str
]) – {"nearest"
,"linear"
,"bilinear"
,"bicubic"
,"trilinear"
} Only used in the “nontrainable” mode. If ends with"linear"
will usespatial 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/generated/torch.nn.Upsample.htmlalign_corners (
Optional
[bool
]) – set the align_corners parameter of torch.nn.Upsample. Defaults to True. Only used in the “nontrainable” mode.bias (
bool
) – whether to have a bias term in the default preconv and deconv layers. Defaults to True.apply_pad_pool (
bool
) – if True the upsampled tensor is padded then average pooling is applied with a kernel the size of scale_factor with a stride of 1. See also:monai.networks.blocks.SubpixelUpsample
. Only used in the “pixelshuffle” mode.
Deprecated since version 0.6.0:
dimensions
is deprecated, usespatial_dims
instead.
- monai.networks.blocks.Upsample¶
- class monai.networks.blocks.SubpixelUpsample(spatial_dims, in_channels, out_channels=None, scale_factor=2, conv_block='default', apply_pad_pool=True, bias=True, dimensions=None)[source]¶
Upsample via using a subpixel CNN. This module supports 1D, 2D and 3D input images. The module is consisted with two parts. First of all, a convolutional layer is employed to increase the number of channels into:
in_channels * (scale_factor ** dimensions)
. Secondly, a pixel shuffle manipulation is utilized to aggregates the feature maps from low resolution space and build the super resolution space. The first part of the module is not fixed, a sequential layers can be used to replace the default single layer.See: Shi et al., 2016, “Real-Time Single Image and Video Super-Resolution Using a nEfficient Sub-Pixel Convolutional Neural Network.”
See: Aitken et al., 2017, “Checkerboard artifact free sub-pixel convolution”.
The idea comes from: https://arxiv.org/abs/1609.05158
The pixel shuffle mechanism refers to: https://pytorch.org/docs/stable/generated/torch.nn.PixelShuffle.html#torch.nn.PixelShuffle. and: https://github.com/pytorch/pytorch/pull/6340.
- Parameters
spatial_dims (
int
) – number of spatial dimensions of the input image.in_channels (
Optional
[int
]) – number of channels of the input image.out_channels (
Optional
[int
]) – optional number of channels of the output image.scale_factor (
int
) – multiplier for spatial size. Defaults to 2.conv_block (
Union
[Module
,str
,None
]) –a conv block to extract feature maps before upsampling. Defaults to None.
When
conv_block
is"default"
, one reserved conv layer will be utilized.When
conv_block
is annn.module
, please ensure the output number of channels is divisible(scale_factor ** dimensions)
.
apply_pad_pool (
bool
) – if True the upsampled tensor is padded then average pooling is applied with a kernel the size of scale_factor with a stride of 1. This implements the nearest neighbour resize convolution component of subpixel convolutions described in Aitken et al.bias (
bool
) – whether to have a bias term in the default conv_block. Defaults to True.
Deprecated since version 0.6.0:
dimensions
is deprecated, usespatial_dims
instead.- __init__(spatial_dims, in_channels, out_channels=None, scale_factor=2, conv_block='default', apply_pad_pool=True, bias=True, dimensions=None)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions of the input image.in_channels (
Optional
[int
]) – number of channels of the input image.out_channels (
Optional
[int
]) – optional number of channels of the output image.scale_factor (
int
) – multiplier for spatial size. Defaults to 2.conv_block (
Union
[Module
,str
,None
]) –a conv block to extract feature maps before upsampling. Defaults to None.
When
conv_block
is"default"
, one reserved conv layer will be utilized.When
conv_block
is annn.module
, please ensure the output number of channels is divisible(scale_factor ** dimensions)
.
apply_pad_pool (
bool
) – if True the upsampled tensor is padded then average pooling is applied with a kernel the size of scale_factor with a stride of 1. This implements the nearest neighbour resize convolution component of subpixel convolutions described in Aitken et al.bias (
bool
) – whether to have a bias term in the default conv_block. Defaults to True.
Deprecated since version 0.6.0:
dimensions
is deprecated, usespatial_dims
instead.
- monai.networks.blocks.Subpixelupsample¶
- monai.networks.blocks.SubpixelUpSample¶
Registration Residual Conv Block¶
- class monai.networks.blocks.RegistrationResidualConvBlock(spatial_dims, in_channels, out_channels, num_layers=2, kernel_size=3)[source]¶
A block with skip links and layer - norm - activation. Only changes the number of channels, the spatial size is kept same.
- Parameters
spatial_dims (
int
) – number of spatial dimensionsin_channels (
int
) – number of input channelsout_channels (
int
) – number of output channelsnum_layers (
int
) – number of layers inside the blockkernel_size (
int
) – kernel_size
- __init__(spatial_dims, in_channels, out_channels, num_layers=2, kernel_size=3)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensionsin_channels (
int
) – number of input channelsout_channels (
int
) – number of output channelsnum_layers (
int
) – number of layers inside the blockkernel_size (
int
) – kernel_size
Registration Down Sample Block¶
- class monai.networks.blocks.RegistrationDownSampleBlock(spatial_dims, channels, pooling)[source]¶
A down-sample module used in RegUNet to half the spatial size. The number of channels is kept same.
- Adapted from:
DeepReg (https://github.com/DeepRegNet/DeepReg)
- Parameters
spatial_dims (
int
) – number of spatial dimensions.channels (
int
) – channelspooling (
bool
) – use MaxPool if True, strided conv if False
- __init__(spatial_dims, channels, pooling)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions.channels (
int
) – channelspooling (
bool
) – use MaxPool if True, strided conv if False
- forward(x)[source]¶
Halves the spatial dimensions and keeps the same channel. output in shape (batch,
channels
, insize_1 / 2, insize_2 / 2, [insize_3 / 2]),- Parameters
x (
Tensor
) – Tensor in shape (batch,channels
, insize_1, insize_2, [insize_3])- Raises
ValueError – when input spatial dimensions are not even.
- Return type
Tensor
Registration Extraction Block¶
- class monai.networks.blocks.RegistrationExtractionBlock(spatial_dims, extract_levels, num_channels, out_channels, kernel_initializer='kaiming_uniform', activation=None)[source]¶
The Extraction Block used in RegUNet. Extracts feature from each
extract_levels
and takes the average.- Parameters
spatial_dims (
int
) – number of spatial dimensionsextract_levels (
Tuple
[int
]) – spatial levels to extract feature from, 0 refers to the input scalenum_channels (
Union
[Tuple
[int
],List
[int
]]) – number of channels at each scale level, List or Tuple of length equals to depth of the RegNetout_channels (
int
) – number of output channelskernel_initializer (
Optional
[str
]) – kernel initializeractivation (
Optional
[str
]) – kernel activation function
- __init__(spatial_dims, extract_levels, num_channels, out_channels, kernel_initializer='kaiming_uniform', activation=None)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensionsextract_levels (
Tuple
[int
]) – spatial levels to extract feature from, 0 refers to the input scalenum_channels (
Union
[Tuple
[int
],List
[int
]]) – number of channels at each scale level, List or Tuple of length equals to depth of the RegNetout_channels (
int
) – number of output channelskernel_initializer (
Optional
[str
]) – kernel initializeractivation (
Optional
[str
]) – kernel activation function
- forward(x, image_size)[source]¶
- Parameters
x (
List
[Tensor
]) – Decoded feature at different spatial levels, sorted from deep to shallowimage_size (
List
[int
]) – output image size
- Return type
Tensor
- Returns
Tensor of shape (batch, out_channels, size1, size2, size3), where (size1, size2, size3) =
image_size
LocalNet DownSample Block¶
- class monai.networks.blocks.LocalNetDownSampleBlock(spatial_dims, in_channels, out_channels, kernel_size)[source]¶
A down-sample module that can be used for LocalNet, based on: Weakly-supervised convolutional neural networks for multimodal image registration. Label-driven weakly-supervised learning for multimodal deformable image registration.
- Adapted from:
DeepReg (https://github.com/DeepRegNet/DeepReg)
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size.
- Raises
NotImplementedError – when
kernel_size
is even
- __init__(spatial_dims, in_channels, out_channels, kernel_size)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size.
- Raises
NotImplementedError – when
kernel_size
is even
- forward(x)[source]¶
Halves the spatial dimensions. A tuple of (x, mid) is returned:
x is the downsample result, in shape (batch,
out_channels
, insize_1 / 2, insize_2 / 2, [insize_3 / 2]),mid is the mid-level feature, in shape (batch,
out_channels
, insize_1, insize_2, [insize_3])
- Parameters
x – Tensor in shape (batch,
in_channels
, insize_1, insize_2, [insize_3])- Raises
ValueError – when input spatial dimensions are not even.
- Return type
Tuple
[Tensor
,Tensor
]
LocalNet UpSample Block¶
- class monai.networks.blocks.LocalNetUpSampleBlock(spatial_dims, in_channels, out_channels)[source]¶
A up-sample module that can be used for LocalNet, based on: Weakly-supervised convolutional neural networks for multimodal image registration. Label-driven weakly-supervised learning for multimodal deformable image registration.
- Adapted from:
DeepReg (https://github.com/DeepRegNet/DeepReg)
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.
- Raises
ValueError – when
in_channels != 2 * out_channels
- __init__(spatial_dims, in_channels, out_channels)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.
- Raises
ValueError – when
in_channels != 2 * out_channels
- forward(x, mid)[source]¶
Halves the channel and doubles the spatial dimensions.
- Parameters
x – feature to be up-sampled, in shape (batch,
in_channels
, insize_1, insize_2, [insize_3])mid – mid-level feature saved during down-sampling, in shape (batch,
out_channels
, midsize_1, midsize_2, [midsize_3])
- Raises
ValueError – when
midsize != insize * 2
- Return type
Tensor
LocalNet Feature Extractor Block¶
- class monai.networks.blocks.LocalNetFeatureExtractorBlock(spatial_dims, in_channels, out_channels, act='RELU', initializer='kaiming_uniform')[source]¶
A feature-extraction module that can be used for LocalNet, based on: Weakly-supervised convolutional neural networks for multimodal image registration. Label-driven weakly-supervised learning for multimodal deformable image registration.
- Adapted from:
DeepReg (https://github.com/DeepRegNet/DeepReg)
Args: spatial_dims: number of spatial dimensions. in_channels: number of input channels. out_channels: number of output channels. act: activation type and arguments. Defaults to ReLU. kernel_initializer: kernel initializer. Defaults to None.
- __init__(spatial_dims, in_channels, out_channels, act='RELU', initializer='kaiming_uniform')[source]¶
Args: spatial_dims: number of spatial dimensions. in_channels: number of input channels. out_channels: number of output channels. act: activation type and arguments. Defaults to ReLU. kernel_initializer: kernel initializer. Defaults to None.
MLP Block¶
- class monai.networks.blocks.MLPBlock(hidden_size, mlp_dim, dropout_rate=0.0)[source]¶
A multi-layer perceptron block, based on: “Dosovitskiy et al., An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale <https://arxiv.org/abs/2010.11929>”
- Parameters
hidden_size (
int
) – dimension of hidden layer.mlp_dim (
int
) – dimension of feedforward layer.dropout_rate (
float
) – faction of the input units to drop.
- __init__(hidden_size, mlp_dim, dropout_rate=0.0)[source]¶
- Parameters
hidden_size (
int
) – dimension of hidden layer.mlp_dim (
int
) – dimension of feedforward layer.dropout_rate (
float
) – faction of the input units to drop.
- forward(x)[source]¶
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.
Patch Embedding Block¶
- class monai.networks.blocks.PatchEmbeddingBlock(in_channels, img_size, patch_size, hidden_size, num_heads, pos_embed, dropout_rate=0.0, spatial_dims=3)[source]¶
A patch embedding block, based on: “Dosovitskiy et al., An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale <https://arxiv.org/abs/2010.11929>”
Example:
>>> from monai.networks.blocks import PatchEmbeddingBlock >>> PatchEmbeddingBlock(in_channels=4, img_size=32, patch_size=8, hidden_size=32, num_heads=4, pos_embed="conv")
- Parameters
in_channels (
int
) – dimension of input channels.img_size (
Union
[Sequence
[int
],int
]) – dimension of input image.patch_size (
Union
[Sequence
[int
],int
]) – dimension of patch size.hidden_size (
int
) – dimension of hidden layer.num_heads (
int
) – number of attention heads.pos_embed (
str
) – position embedding layer type.dropout_rate (
float
) – faction of the input units to drop.spatial_dims (
int
) – number of spatial dimensions.
- __init__(in_channels, img_size, patch_size, hidden_size, num_heads, pos_embed, dropout_rate=0.0, spatial_dims=3)[source]¶
- Parameters
in_channels (
int
) – dimension of input channels.img_size (
Union
[Sequence
[int
],int
]) – dimension of input image.patch_size (
Union
[Sequence
[int
],int
]) – dimension of patch size.hidden_size (
int
) – dimension of hidden layer.num_heads (
int
) – number of attention heads.pos_embed (
str
) – position embedding layer type.dropout_rate (
float
) – faction of the input units to drop.spatial_dims (
int
) – number of spatial dimensions.
- forward(x)[source]¶
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.
FactorizedIncreaseBlock¶
- class monai.networks.blocks.FactorizedIncreaseBlock(in_channel, out_channel, spatial_dims=3, act_name='RELU', norm_name='INSTANCE')[source]¶
Up-sampling the features by two using linear interpolation and convolutions.
- Parameters
in_channel (
int
) – number of input channelsout_channel (
int
) – number of output channelsspatial_dims (
int
) – number of spatial dimensionsact_name (
Union
[Tuple
,str
]) – activation layer type and arguments.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.
- __init__(in_channel, out_channel, spatial_dims=3, act_name='RELU', norm_name='INSTANCE')[source]¶
- Parameters
in_channel (
int
) – number of input channelsout_channel (
int
) – number of output channelsspatial_dims (
int
) – number of spatial dimensionsact_name (
Union
[Tuple
,str
]) – activation layer type and arguments.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.
FactorizedReduceBlock¶
- class monai.networks.blocks.FactorizedReduceBlock(in_channel, out_channel, spatial_dims=3, act_name='RELU', norm_name='INSTANCE')[source]¶
Down-sampling the feature by 2 using stride. The length along each spatial dimension must be a multiple of 2.
- Parameters
in_channel (
int
) – number of input channelsout_channel (
int
) – number of output channels.spatial_dims (
int
) – number of spatial dimensions.act_name (
Union
[Tuple
,str
]) – activation layer type and arguments.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.
- __init__(in_channel, out_channel, spatial_dims=3, act_name='RELU', norm_name='INSTANCE')[source]¶
- Parameters
in_channel (
int
) – number of input channelsout_channel (
int
) – number of output channels.spatial_dims (
int
) – number of spatial dimensions.act_name (
Union
[Tuple
,str
]) – activation layer type and arguments.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.
P3DActiConvNormBlock¶
- class monai.networks.blocks.P3DActiConvNormBlock(in_channel, out_channel, kernel_size, padding, mode=0, act_name='RELU', norm_name='INSTANCE')[source]¶
– (act) – (conv) – (norm) –
- Parameters
in_channel (
int
) – number of input channels.out_channel (
int
) – number of output channels.kernel_size (
int
) – kernel size to be expanded to 3D.padding (
int
) – padding size to be expanded to 3D.mode (
int
) –mode for the anisotropic kernels:
0:
(k, k, 1)
,(1, 1, k)
,1:
(k, 1, k)
,(1, k, 1)
,2:
(1, k, k)
.(k, 1, 1)
.
act_name (
Union
[Tuple
,str
]) – activation layer type and arguments.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.
- __init__(in_channel, out_channel, kernel_size, padding, mode=0, act_name='RELU', norm_name='INSTANCE')[source]¶
- Parameters
in_channel (
int
) – number of input channels.out_channel (
int
) – number of output channels.kernel_size (
int
) – kernel size to be expanded to 3D.padding (
int
) – padding size to be expanded to 3D.mode (
int
) –mode for the anisotropic kernels:
0:
(k, k, 1)
,(1, 1, k)
,1:
(k, 1, k)
,(1, k, 1)
,2:
(1, k, k)
.(k, 1, 1)
.
act_name (
Union
[Tuple
,str
]) – activation layer type and arguments.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.
ActiConvNormBlock¶
- class monai.networks.blocks.ActiConvNormBlock(in_channel, out_channel, kernel_size=3, padding=1, spatial_dims=3, act_name='RELU', norm_name='INSTANCE')[source]¶
– (Acti) – (Conv) – (Norm) –
- Parameters
in_channel (
int
) – number of input channels.out_channel (
int
) – number of output channels.kernel_size (
int
) – kernel size of the convolution.padding (
int
) – padding size of the convolution.spatial_dims (
int
) – number of spatial dimensions.act_name (
Union
[Tuple
,str
]) – activation layer type and arguments.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.
- __init__(in_channel, out_channel, kernel_size=3, padding=1, spatial_dims=3, act_name='RELU', norm_name='INSTANCE')[source]¶
- Parameters
in_channel (
int
) – number of input channels.out_channel (
int
) – number of output channels.kernel_size (
int
) – kernel size of the convolution.padding (
int
) – padding size of the convolution.spatial_dims (
int
) – number of spatial dimensions.act_name (
Union
[Tuple
,str
]) – activation layer type and arguments.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.
Warp¶
- class monai.networks.blocks.Warp(mode='bilinear', padding_mode='border')[source]¶
Warp an image with given dense displacement field (DDF).
For pytorch native APIs, the possible values are:
mode:
"nearest"
,"bilinear"
,"bicubic"
.padding_mode:
"zeros"
,"border"
,"reflection"
See also: https://pytorch.org/docs/stable/generated/torch.nn.functional.grid_sample.html
For MONAI C++/CUDA extensions, the possible values are:
mode:
"nearest"
,"bilinear"
,"bicubic"
, 0, 1, …padding_mode:
"zeros"
,"border"
,"reflection"
, 0, 1, …
See also:
monai.networks.layers.grid_pull
- __init__(mode='bilinear', padding_mode='border')[source]¶
For pytorch native APIs, the possible values are:
mode:
"nearest"
,"bilinear"
,"bicubic"
.padding_mode:
"zeros"
,"border"
,"reflection"
See also: https://pytorch.org/docs/stable/generated/torch.nn.functional.grid_sample.html
For MONAI C++/CUDA extensions, the possible values are:
mode:
"nearest"
,"bilinear"
,"bicubic"
, 0, 1, …padding_mode:
"zeros"
,"border"
,"reflection"
, 0, 1, …
See also:
monai.networks.layers.grid_pull
DVF2DDF¶
- class monai.networks.blocks.DVF2DDF(num_steps=7, mode='bilinear', padding_mode='zeros')[source]¶
Layer calculates a dense displacement field (DDF) from a dense velocity field (DVF) with scaling and squaring.
- Adapted from:
DeepReg (https://github.com/DeepRegNet/DeepReg)
Initializes internal Module state, shared by both nn.Module and ScriptModule.
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 normalization 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.LayerFactory[source]¶
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)[source]¶
Add the factory function to this object under the given name.
- Return type
None
- factory_function(name)[source]¶
Decorator for adding a factory function with the given name.
- Return type
Callable
- get_constructor(factory_name, *args)[source]¶
Get the constructor for the given factory name and arguments.
- Raises
TypeError – When
factory_name
is not astr
.- Return type
Any
- property names: Tuple[str, ...]¶
Produces all factory names.
- Return type
Tuple
[str
, …]
split_args¶
- monai.networks.layers.split_args(args)[source]¶
Split arguments in a way to be suitable for using with the factory types. If args is a string it’s interpreted as the type name.
- Parameters
args (str or a tuple of object name and kwarg dict) – input arguments to be parsed.
- Raises
TypeError – When
args
type is not inUnion[str, Tuple[Union[str, Callable], dict]]
.
Examples:
>>> act_type, args = split_args("PRELU") >>> monai.networks.layers.Act[act_type] <class 'torch.nn.modules.activation.PReLU'> >>> act_type, args = split_args(("PRELU", {"num_parameters": 1, "init": 0.25})) >>> monai.networks.layers.Act[act_type](**args) PReLU(num_parameters=1)
Dropout¶
The supported members are: DROPOUT
, ALPHADROPOUT
.
Please see monai.networks.layers.split_args
for additional args parsing.
Act¶
The supported members are: ELU
, RELU
, LEAKYRELU
, PRELU
, RELU6
, SELU
, CELU
, GELU
, SIGMOID
, TANH
, SOFTMAX
, LOGSOFTMAX
, SWISH
, MEMSWISH
, MISH
.
Please see monai.networks.layers.split_args
for additional args parsing.
Norm¶
The supported members are: INSTANCE
, BATCH
, GROUP
, LAYER
, LOCALRESPONSE
, SYNCBATCH
.
Please see monai.networks.layers.split_args
for additional args parsing.
Conv¶
The supported members are: CONV
, CONVTRANS
.
Please see monai.networks.layers.split_args
for additional args parsing.
Pad¶
The supported members are: REPLICATIONPAD
, CONSTANTPAD
.
Please see monai.networks.layers.split_args
for additional args parsing.
Pool¶
The supported members are: MAX
, ADAPTIVEMAX
, AVG
, ADAPTIVEAVG
.
Please see monai.networks.layers.split_args
for additional args parsing.
ChannelPad¶
- class monai.networks.layers.ChannelPad(spatial_dims, in_channels, out_channels, mode=ChannelMatching.PAD)[source]¶
Expand the input tensor’s channel dimension from length in_channels to out_channels, by padding or a projection.
- 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.mode (
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 one.
- __init__(spatial_dims, in_channels, out_channels, mode=ChannelMatching.PAD)[source]¶
- 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.mode (
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 one.
- forward(x)[source]¶
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.- Return type
Tensor
SkipConnection¶
- class monai.networks.layers.SkipConnection(submodule, dim=1, mode='cat')[source]¶
Combine the forward pass input with the result from the given submodule:
--+--submodule--o-- |_____________|
The available modes are
"cat"
,"add"
,"mul"
.- Parameters
submodule – the module defines the trainable branch.
dim (
int
) – the dimension over which the tensors are concatenated. Used when mode is"cat"
.mode (
Union
[str
,SkipMode
]) –"cat"
,"add"
,"mul"
. defaults to"cat"
.
- __init__(submodule, dim=1, mode='cat')[source]¶
- Parameters
submodule – the module defines the trainable branch.
dim (
int
) – the dimension over which the tensors are concatenated. Used when mode is"cat"
.mode (
Union
[str
,SkipMode
]) –"cat"
,"add"
,"mul"
. defaults to"cat"
.
- forward(x)[source]¶
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.- Return type
Tensor
Flatten¶
- class monai.networks.layers.Flatten[source]¶
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)[source]¶
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.- Return type
Tensor
Reshape¶
- class monai.networks.layers.Reshape(*shape)[source]¶
Reshapes input tensors to the given shape (minus batch dimension), retaining original batch size.
Given a shape list/tuple shape of integers (s0, s1, … , sn), this layer will reshape input tensors of shape (batch, s0 * s1 * … * sn) to shape (batch, s0, s1, … , sn).
- Parameters
shape (
int
) – list/tuple of integer shape dimensions
- __init__(*shape)[source]¶
Given a shape list/tuple shape of integers (s0, s1, … , sn), this layer will reshape input tensors of shape (batch, s0 * s1 * … * sn) to shape (batch, s0, s1, … , sn).
- Parameters
shape (
int
) – list/tuple of integer shape dimensions
- forward(x)[source]¶
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.- Return type
Tensor
separable_filtering¶
- monai.networks.layers.separable_filtering(x, kernels, mode='zeros')[source]¶
Apply 1-D convolutions along each spatial dimension of x.
- Parameters
x (
Tensor
) – the input image. must have shape (batch, channels, H[, W, …]).kernels (
List
[Tensor
]) – kernel along each spatial dimension. could be a single kernel (duplicated for all spatial dimensions), or a list of spatial_dims number of kernels.mode (string, optional) – padding mode passed to convolution class.
'zeros'
,'reflect'
,'replicate'
or'circular'
. Default:'zeros'
. Seetorch.nn.Conv1d()
for more information.
- Raises
TypeError – When
x
is not atorch.Tensor
.
Examples:
>>> import torch >>> from monai.networks.layers import separable_filtering >>> img = torch.randn(2, 4, 32, 32) # batch_size 2, channels 4, 32x32 2D images # applying a [-1, 0, 1] filter along each of the spatial dimensions. # the output shape is the same as the input shape. >>> out = separable_filtering(img, torch.tensor((-1., 0., 1.))) # applying `[-1, 0, 1]`, `[1, 0, -1]` filters along two spatial dimensions respectively. # the output shape is the same as the input shape. >>> out = separable_filtering(img, [torch.tensor((-1., 0., 1.)), torch.tensor((1., 0., -1.))])
- Return type
Tensor
apply_filter¶
- monai.networks.layers.apply_filter(x, kernel, **kwargs)[source]¶
Filtering x with kernel independently for each batch and channel respectively.
- Parameters
x (
Tensor
) – the input image, must have shape (batch, channels, H[, W, D]).kernel (
Tensor
) – kernel must at least have the spatial shape (H_k[, W_k, D_k]). kernel shape must be broadcastable to the batch and channels dimensions of x.kwargs – keyword arguments passed to conv*d() functions.
- Return type
Tensor
- Returns
The filtered x.
Examples:
>>> import torch >>> from monai.networks.layers import apply_filter >>> img = torch.rand(2, 5, 10, 10) # batch_size 2, channels 5, 10x10 2D images >>> out = apply_filter(img, torch.rand(3, 3)) # spatial kernel >>> out = apply_filter(img, torch.rand(5, 3, 3)) # channel-wise kernels >>> out = apply_filter(img, torch.rand(2, 5, 3, 3)) # batch-, channel-wise kernels
GaussianFilter¶
- class monai.networks.layers.GaussianFilter(spatial_dims, sigma, truncated=4.0, approx='erf', requires_grad=False)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions of the input image. must have shape (Batch, channels, H[, W, …]).sigma (
Union
[Sequence
[float
],float
,Sequence
[Tensor
],Tensor
]) – std. could be a single value, or spatial_dims number of values.truncated (
float
) – spreads how many stds.approx (
str
) –discrete Gaussian kernel type, available options are “erf”, “sampled”, and “scalespace”.
erf
approximation interpolates the error function;sampled
uses a sampled Gaussian kernel;scalespace
corresponds to https://en.wikipedia.org/wiki/Scale_space_implementation#The_discrete_Gaussian_kernel based on the modified Bessel functions.
requires_grad (
bool
) – whether to store the gradients for sigma. if True, sigma will be the initial value of the parameters of this module (for example parameters() iterator could be used to get the parameters); otherwise this module will fix the kernels using sigma as the std.
- __init__(spatial_dims, sigma, truncated=4.0, approx='erf', requires_grad=False)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimensions of the input image. must have shape (Batch, channels, H[, W, …]).sigma (
Union
[Sequence
[float
],float
,Sequence
[Tensor
],Tensor
]) – std. could be a single value, or spatial_dims number of values.truncated (
float
) – spreads how many stds.approx (
str
) –discrete Gaussian kernel type, available options are “erf”, “sampled”, and “scalespace”.
erf
approximation interpolates the error function;sampled
uses a sampled Gaussian kernel;scalespace
corresponds to https://en.wikipedia.org/wiki/Scale_space_implementation#The_discrete_Gaussian_kernel based on the modified Bessel functions.
requires_grad (
bool
) – whether to store the gradients for sigma. if True, sigma will be the initial value of the parameters of this module (for example parameters() iterator could be used to get the parameters); otherwise this module will fix the kernels using sigma as the std.
BilateralFilter¶
- class monai.networks.layers.BilateralFilter(*args, **kwargs)[source]¶
Blurs the input tensor spatially whilst preserving edges. Can run on 1D, 2D, or 3D, tensors (on top of Batch and Channel dimensions). Two implementations are provided, an exact solution and a much faster approximation which uses a permutohedral lattice.
- See:
https://en.wikipedia.org/wiki/Bilateral_filter https://graphics.stanford.edu/papers/permutohedral/
- Parameters
input – input tensor.
sigma (color) – the standard deviation of the spatial blur. Higher values can hurt performance when not using the approximate method (see fast approx).
sigma – the standard deviation of the color blur. Lower values preserve edges better whilst higher values tend to a simple gaussian spatial blur.
approx (fast) – This flag chooses between two implementations. The approximate method may produce artifacts in some scenarios whereas the exact solution may be intolerably slow for high spatial standard deviations.
- Returns
output tensor.
- Return type
output (torch.Tensor)
- static backward(ctx, grad_output)[source]¶
Defines a formula for differentiating the operation with backward mode automatic differentiation (alias to the vjp function).
This function is to be overridden by all subclasses.
It must accept a context
ctx
as the first argument, followed by as many outputs as theforward()
returned (None will be passed in for non tensor outputs of the forward function), and it should return as many tensors, as there were inputs toforward()
. Each argument is the gradient w.r.t the given output, and each returned value should be the gradient w.r.t. the corresponding input. If an input is not a Tensor or is a Tensor not requiring grads, you can just pass None as a gradient for that input.The context can be used to retrieve tensors saved during the forward pass. It also has an attribute
ctx.needs_input_grad
as a tuple of booleans representing whether each input needs gradient. E.g.,backward()
will havectx.needs_input_grad[0] = True
if the first input toforward()
needs gradient computated w.r.t. the output.
- static forward(ctx, input, spatial_sigma=5, color_sigma=0.5, fast_approx=True)[source]¶
Performs the operation.
This function is to be overridden by all subclasses.
It must accept a context ctx as the first argument, followed by any number of arguments (tensors or other types).
The context can be used to store arbitrary data that can be then retrieved during the backward pass. Tensors should not be stored directly on ctx (though this is not currently enforced for backward compatibility). Instead, tensors should be saved either with
ctx.save_for_backward()
if they are intended to be used inbackward
(equivalently,vjp
) orctx.save_for_forward()
if they are intended to be used for injvp
.
PHLFilter¶
- class monai.networks.layers.PHLFilter(*args, **kwargs)[source]¶
Filters input based on arbitrary feature vectors. Uses a permutohedral lattice data structure to efficiently approximate n-dimensional gaussian filtering. Complexity is broadly independent of kernel size. Most applicable to higher filter dimensions and larger kernel sizes.
- Parameters
input – input tensor to be filtered.
features – feature tensor used to filter the input.
sigmas – the standard deviations of each feature in the filter.
- Returns
output tensor.
- Return type
output (torch.Tensor)
GaussianMixtureModel¶
- class monai.networks.layers.GaussianMixtureModel(channel_count, mixture_count, mixture_size, verbose_build=False)[source]¶
Takes an initial labeling and uses a mixture of Gaussians to approximate each classes distribution in the feature space. Each unlabeled element is then assigned a probability of belonging to each class based on it’s fit to each classes approximated distribution.
- Parameters
channel_count (
int
) – The number of features per element.mixture_count (
int
) – The number of class distributions.mixture_size (
int
) – The number Gaussian components per class distribution.verbose_build (
bool
) – IfTrue
, turns on verbose logging of load steps.
SavitzkyGolayFilter¶
- class monai.networks.layers.SavitzkyGolayFilter(window_length, order, axis=2, mode='zeros')[source]¶
Convolve a Tensor along a particular axis with a Savitzky-Golay kernel.
- Parameters
window_length (
int
) – Length of the filter window, must be a positive odd integer.order (
int
) – Order of the polynomial to fit to each window, must be less thanwindow_length
.axis (optional) – Axis along which to apply the filter kernel. Default 2 (first spatial dimension).
mode (string, optional) – padding mode passed to convolution class.
'zeros'
,'reflect'
,'replicate'
orDefault ('circular'.) –
'zeros'
. See torch.nn.Conv1d() for more information.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
- Parameters
x (
Tensor
) – Tensor or array-like to filter. Must be real, in shape[Batch, chns, spatial1, spatial2, ...]
and have a device type of'cpu'
.- Returns
x
filtered by Savitzky-Golay kernel with window lengthself.window_length
using polynomials of orderself.order
, along axis specified inself.axis
.- Return type
torch.Tensor
HilbertTransform¶
- class monai.networks.layers.HilbertTransform(axis=2, n=None)[source]¶
Determine the analytical signal of a Tensor along a particular axis. Requires PyTorch 1.7.0+ and the PyTorch FFT module (which is not included in NVIDIA PyTorch Release 20.10).
- Parameters
axis (
int
) – Axis along which to apply Hilbert transform. Default 2 (first spatial dimension).n (
Optional
[int
]) – Number of Fourier components (i.e. FFT size). Default:x.shape[axis]
.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
- Parameters
x (
Tensor
) – Tensor or array-like to transform. Must be real and in shape[Batch, chns, spatial1, spatial2, ...]
.- Returns
Analytical signal of
x
, transformed along axis specified inself.axis
using FFT of sizeself.N
. The absolute value ofx_ht
relates to the envelope ofx
along axisself.axis
.- Return type
torch.Tensor
Affine Transform¶
- class monai.networks.layers.AffineTransform(spatial_size=None, normalized=False, mode=GridSampleMode.BILINEAR, padding_mode=GridSamplePadMode.ZEROS, align_corners=False, reverse_indexing=True)[source]¶
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) inputmatrix
ofscipy.ndimage.affine_transform
, operates on homogeneous coordinates. See also: https://docs.scipy.org/doc/scipy/reference/generated/scipy.ndimage.affine_transform.htmlWhen 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 (
Union
[Sequence
[int
],int
,None
]) – 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/generated/torch.nn.functional.grid_sample.htmlpadding_mode (
Union
[GridSamplePadMode
,str
]) – {"zeros"
,"border"
,"reflection"
} Padding mode for outside grid values. Defaults to"zeros"
. See also: https://pytorch.org/docs/stable/generated/torch.nn.functional.grid_sample.htmlalign_corners (
bool
) – see also https://pytorch.org/docs/stable/generated/torch.nn.functional.grid_sample.html.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.
- __init__(spatial_size=None, normalized=False, mode=GridSampleMode.BILINEAR, padding_mode=GridSamplePadMode.ZEROS, align_corners=False, reverse_indexing=True)[source]¶
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) inputmatrix
ofscipy.ndimage.affine_transform
, operates on homogeneous coordinates. See also: https://docs.scipy.org/doc/scipy/reference/generated/scipy.ndimage.affine_transform.htmlWhen 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 (
Union
[Sequence
[int
],int
,None
]) – 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/generated/torch.nn.functional.grid_sample.htmlpadding_mode (
Union
[GridSamplePadMode
,str
]) – {"zeros"
,"border"
,"reflection"
} Padding mode for outside grid values. Defaults to"zeros"
. See also: https://pytorch.org/docs/stable/generated/torch.nn.functional.grid_sample.htmlalign_corners (
bool
) – see also https://pytorch.org/docs/stable/generated/torch.nn.functional.grid_sample.html.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)[source]¶
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 (
Union
[Sequence
[int
],int
,None
]) – output spatial shape, the full output shape will be [N, C, *spatial_size] where N and C are inferred from the src.
- Raises
TypeError – When
theta
is not atorch.Tensor
.ValueError – When
theta
is not one of [Nxdxd, dxd].ValueError – When
theta
is not one of [Nx3x3, Nx4x4].TypeError – When
src
is not atorch.Tensor
.ValueError – When
src
spatially is not one of [2D, 3D].ValueError – When affine and image batch dimension differ.
- Return type
Tensor
grid_pull¶
- monai.networks.layers.grid_pull(input, grid, interpolation='linear', bound='zero', extrapolate=True)[source]¶
Sample an image with respect to a deformation field.
interpolation can be an int, a string or an InterpolationType. Possible values are:
- 0 or 'nearest' or InterpolationType.nearest - 1 or 'linear' or InterpolationType.linear - 2 or 'quadratic' or InterpolationType.quadratic - 3 or 'cubic' or InterpolationType.cubic - 4 or 'fourth' or InterpolationType.fourth - 5 or 'fifth' or InterpolationType.fifth - 6 or 'sixth' or InterpolationType.sixth - 7 or 'seventh' or InterpolationType.seventh
A list of values can be provided, in the order [W, H, D], to specify dimension-specific interpolation orders.
bound can be an int, a string or a BoundType. Possible values are:
- 0 or 'replicate' or 'nearest' or BoundType.replicate - 1 or 'dct1' or 'mirror' or BoundType.dct1 - 2 or 'dct2' or 'reflect' or BoundType.dct2 - 3 or 'dst1' or 'antimirror' or BoundType.dst1 - 4 or 'dst2' or 'antireflect' or BoundType.dst2 - 5 or 'dft' or 'wrap' or BoundType.dft - 7 or 'zero' or BoundType.zero
A list of values can be provided, in the order [W, H, D], to specify dimension-specific boundary conditions. sliding is a specific condition than only applies to flow fields (with as many channels as dimensions). It cannot be dimension-specific. Note that:
dft corresponds to circular padding
dct2 corresponds to Neumann boundary conditions (symmetric)
dst2 corresponds to Dirichlet boundary conditions (antisymmetric)
See also
help(monai._C.BoundType)
help(monai._C.InterpolationType)
- Parameters
input (
Tensor
) – Input image. (B, C, Wi, Hi, Di).grid (
Tensor
) – Deformation field. (B, Wo, Ho, Do, 1|2|3).interpolation (int or list[int] , optional) – Interpolation order. Defaults to ‘linear’.
bound (BoundType, or list[BoundType], optional) – Boundary conditions. Defaults to ‘zero’.
extrapolate (
bool
) – Extrapolate out-of-bound data. Defaults to True.
- Returns
Deformed image (B, C, Wo, Ho, Do).
- Return type
output (torch.Tensor)
grid_push¶
- monai.networks.layers.grid_push(input, grid, shape=None, interpolation='linear', bound='zero', extrapolate=True)[source]¶
Splat an image with respect to a deformation field (pull adjoint).
interpolation can be an int, a string or an InterpolationType. Possible values are:
- 0 or 'nearest' or InterpolationType.nearest - 1 or 'linear' or InterpolationType.linear - 2 or 'quadratic' or InterpolationType.quadratic - 3 or 'cubic' or InterpolationType.cubic - 4 or 'fourth' or InterpolationType.fourth - 5 or 'fifth' or InterpolationType.fifth - 6 or 'sixth' or InterpolationType.sixth - 7 or 'seventh' or InterpolationType.seventh
A list of values can be provided, in the order [W, H, D], to specify dimension-specific interpolation orders.
bound can be an int, a string or a BoundType. Possible values are:
- 0 or 'replicate' or 'nearest' or BoundType.replicate - 1 or 'dct1' or 'mirror' or BoundType.dct1 - 2 or 'dct2' or 'reflect' or BoundType.dct2 - 3 or 'dst1' or 'antimirror' or BoundType.dst1 - 4 or 'dst2' or 'antireflect' or BoundType.dst2 - 5 or 'dft' or 'wrap' or BoundType.dft - 7 or 'zero' or BoundType.zero
A list of values can be provided, in the order [W, H, D], to specify dimension-specific boundary conditions. sliding is a specific condition than only applies to flow fields (with as many channels as dimensions). It cannot be dimension-specific. Note that:
dft corresponds to circular padding
dct2 corresponds to Neumann boundary conditions (symmetric)
dst2 corresponds to Dirichlet boundary conditions (antisymmetric)
See also
help(monai._C.BoundType)
help(monai._C.InterpolationType)
- Parameters
input (
Tensor
) – Input image (B, C, Wi, Hi, Di).grid (
Tensor
) – Deformation field (B, Wi, Hi, Di, 1|2|3).shape – Shape of the source image.
interpolation (int or list[int] , optional) – Interpolation order. Defaults to ‘linear’.
bound (BoundType, or list[BoundType], optional) – Boundary conditions. Defaults to ‘zero’.
extrapolate (
bool
) – Extrapolate out-of-bound data. Defaults to True.
- Returns
Splatted image (B, C, Wo, Ho, Do).
- Return type
output (torch.Tensor)
grid_count¶
- monai.networks.layers.grid_count(grid, shape=None, interpolation='linear', bound='zero', extrapolate=True)[source]¶
Splatting weights with respect to a deformation field (pull adjoint).
This function is equivalent to applying grid_push to an image of ones.
interpolation can be an int, a string or an InterpolationType. Possible values are:
- 0 or 'nearest' or InterpolationType.nearest - 1 or 'linear' or InterpolationType.linear - 2 or 'quadratic' or InterpolationType.quadratic - 3 or 'cubic' or InterpolationType.cubic - 4 or 'fourth' or InterpolationType.fourth - 5 or 'fifth' or InterpolationType.fifth - 6 or 'sixth' or InterpolationType.sixth - 7 or 'seventh' or InterpolationType.seventh
A list of values can be provided, in the order [W, H, D], to specify dimension-specific interpolation orders.
bound can be an int, a string or a BoundType. Possible values are:
- 0 or 'replicate' or 'nearest' or BoundType.replicate - 1 or 'dct1' or 'mirror' or BoundType.dct1 - 2 or 'dct2' or 'reflect' or BoundType.dct2 - 3 or 'dst1' or 'antimirror' or BoundType.dst1 - 4 or 'dst2' or 'antireflect' or BoundType.dst2 - 5 or 'dft' or 'wrap' or BoundType.dft - 7 or 'zero' or BoundType.zero
A list of values can be provided, in the order [W, H, D], to specify dimension-specific boundary conditions. sliding is a specific condition than only applies to flow fields (with as many channels as dimensions). It cannot be dimension-specific. Note that:
dft corresponds to circular padding
dct2 corresponds to Neumann boundary conditions (symmetric)
dst2 corresponds to Dirichlet boundary conditions (antisymmetric)
See also
help(monai._C.BoundType)
help(monai._C.InterpolationType)
- Parameters
grid (
Tensor
) – Deformation field (B, Wi, Hi, Di, 2|3).shape – shape of the source image.
interpolation (int or list[int] , optional) – Interpolation order. Defaults to ‘linear’.
bound (BoundType, or list[BoundType], optional) – Boundary conditions. Defaults to ‘zero’.
extrapolate (bool, optional) – Extrapolate out-of-bound data. Defaults to True.
- Returns
Splat weights (B, 1, Wo, Ho, Do).
- Return type
output (torch.Tensor)
grid_grad¶
- monai.networks.layers.grid_grad(input, grid, interpolation='linear', bound='zero', extrapolate=True)[source]¶
Sample an image with respect to a deformation field.
interpolation can be an int, a string or an InterpolationType. Possible values are:
- 0 or 'nearest' or InterpolationType.nearest - 1 or 'linear' or InterpolationType.linear - 2 or 'quadratic' or InterpolationType.quadratic - 3 or 'cubic' or InterpolationType.cubic - 4 or 'fourth' or InterpolationType.fourth - 5 or 'fifth' or InterpolationType.fifth - 6 or 'sixth' or InterpolationType.sixth - 7 or 'seventh' or InterpolationType.seventh
A list of values can be provided, in the order [W, H, D], to specify dimension-specific interpolation orders.
bound can be an int, a string or a BoundType. Possible values are:
- 0 or 'replicate' or 'nearest' or BoundType.replicate - 1 or 'dct1' or 'mirror' or BoundType.dct1 - 2 or 'dct2' or 'reflect' or BoundType.dct2 - 3 or 'dst1' or 'antimirror' or BoundType.dst1 - 4 or 'dst2' or 'antireflect' or BoundType.dst2 - 5 or 'dft' or 'wrap' or BoundType.dft - 7 or 'zero' or BoundType.zero
A list of values can be provided, in the order [W, H, D], to specify dimension-specific boundary conditions. sliding is a specific condition than only applies to flow fields (with as many channels as dimensions). It cannot be dimension-specific. Note that:
dft corresponds to circular padding
dct2 corresponds to Neumann boundary conditions (symmetric)
dst2 corresponds to Dirichlet boundary conditions (antisymmetric)
See also
help(monai._C.BoundType)
help(monai._C.InterpolationType)
- Parameters
input (
Tensor
) – Input image. (B, C, Wi, Hi, Di).grid (
Tensor
) – Deformation field. (B, Wo, Ho, Do, 2|3).interpolation (int or list[int] , optional) – Interpolation order. Defaults to ‘linear’.
bound (BoundType, or list[BoundType], optional) – Boundary conditions. Defaults to ‘zero’.
extrapolate (
bool
) – Extrapolate out-of-bound data. Defaults to True.
- Returns
Sampled gradients (B, C, Wo, Ho, Do, 1|2|3).
- Return type
output (torch.Tensor)
LLTM¶
- class monai.networks.layers.LLTM(input_features, state_size)[source]¶
This recurrent unit is similar to an LSTM, but differs in that it lacks a forget gate and uses an Exponential Linear Unit (ELU) as its internal activation function. Because this unit never forgets, call it LLTM, or Long-Long-Term-Memory unit. It has both C++ and CUDA implementation, automatically switch according to the target device where put this module to.
- Parameters
input_features (
int
) – size of input feature datastate_size (
int
) – size of the state of recurrent unit
Referring to: https://pytorch.org/tutorials/advanced/cpp_extension.html
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(input, state)[source]¶
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¶
- monai.networks.layers.convutils.calculate_out_shape(in_shape, kernel_size, stride, padding)[source]¶
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.
- Return type
Union
[Tuple
[int
, …],int
]
- monai.networks.layers.convutils.gaussian_1d(sigma, truncated=4.0, approx='erf', normalize=False)[source]¶
one dimensional Gaussian kernel.
- Parameters
sigma (
Tensor
) – std of the kerneltruncated (
float
) – tail lengthapprox (
str
) –discrete Gaussian kernel type, available options are “erf”, “sampled”, and “scalespace”.
erf
approximation interpolates the error function;sampled
uses a sampled Gaussian kernel;scalespace
corresponds to https://en.wikipedia.org/wiki/Scale_space_implementation#The_discrete_Gaussian_kernel based on the modified Bessel functions.
normalize (
bool
) – whether to normalize the kernel with kernel.sum().
- Raises
ValueError – When
truncated
is non-positive.- Return type
Tensor
- Returns
1D torch tensor
- monai.networks.layers.convutils.polyval(coef, x)[source]¶
Evaluates the polynomial defined by coef at x.
For a 1D sequence of coef (length n), evaluate:
y = coef[n-1] + x * (coef[n-2] + ... + x * (coef[1] + x * coef[0]))
- Parameters
coef – a sequence of floats representing the coefficients of the polynomial
x – float or a sequence of floats representing the variable of the polynomial
- Return type
Tensor
- Returns
1D torch tensor
- monai.networks.layers.convutils.same_padding(kernel_size, dilation=1)[source]¶
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 – When
np.any((kernel_size - 1) * dilation % 2 == 1)
.- Return type
Union
[Tuple
[int
, …],int
]
- monai.networks.layers.utils.get_act_layer(name)[source]¶
Create an activation layer instance.
For example, to create activation layers:
from monai.networks.layers import get_act_layer s_layer = get_act_layer(name="swish") p_layer = get_act_layer(name=("prelu", {"num_parameters": 1, "init": 0.25}))
- Parameters
name (
Union
[Tuple
,str
]) – an activation type string or a tuple of type string and parameters.
- monai.networks.layers.utils.get_dropout_layer(name, dropout_dim=1)[source]¶
Create a dropout layer instance.
For example, to create dropout layers:
from monai.networks.layers import get_dropout_layer d_layer = get_dropout_layer(name="dropout") a_layer = get_dropout_layer(name=("alphadropout", {"p": 0.25}))
- Parameters
name (
Union
[Tuple
,str
,float
,int
]) – a dropout ratio or a tuple of dropout type and parameters.dropout_dim (
Optional
[int
]) – the spatial dimension of the dropout operation.
- monai.networks.layers.utils.get_norm_layer(name, spatial_dims=1, channels=1)[source]¶
Create a normalization layer instance.
For example, to create normalization layers:
from monai.networks.layers import get_norm_layer g_layer = get_norm_layer(name=("group", {"num_groups": 1})) n_layer = get_norm_layer(name="instance", spatial_dims=2)
- Parameters
name (
Union
[Tuple
,str
]) – a normalization type string or a tuple of type string and parameters.spatial_dims (
Optional
[int
]) – number of spatial dimensions of the input.channels (
Optional
[int
]) – number of features/channels when the normalization layer requires this parameter but it is not specified in the norm parameters.
- monai.networks.layers.utils.get_pool_layer(name, spatial_dims=1)[source]¶
Create a pooling layer instance.
For example, to create adaptiveavg layer:
from monai.networks.layers import get_pool_layer pool_layer = get_pool_layer(("adaptiveavg", {"output_size": (1, 1, 1)}), spatial_dims=3)
- Parameters
name (
Union
[Tuple
,str
]) – a pooling type string or a tuple of type string and parameters.spatial_dims (
Optional
[int
]) – number of spatial dimensions of the input.
Nets¶
AHNet¶
- class monai.networks.nets.AHNet(layers=(3, 4, 6, 3), spatial_dims=3, in_channels=1, out_channels=1, psp_block_num=4, upsample_mode='transpose', pretrained=False, progress=True)[source]¶
AHNet based on Anisotropic Hybrid Network. Adapted from lsqshr’s official code. Except from the original network that supports 3D inputs, this implementation also supports 2D inputs. According to the tests for deconvolutions, using
"transpose"
rather than linear interpolations is faster. Therefore, this implementation sets"transpose"
as the default upsampling method.To meet the requirements of the structure, the input size for each spatial dimension (except the last one) should be: divisible by 2 ** (psp_block_num + 3) and no less than 32 in
transpose
mode, and should be divisible by 32 and no less than 2 ** (psp_block_num + 3) in other upsample modes. In addition, the input size for the last spatial dimension should be divisible by 32, and at least one spatial size should be no less than 64.- Parameters
layers (
tuple
) – number of residual blocks for 4 layers of the network (layer1…layer4). Defaults to(3, 4, 6, 3)
.spatial_dims (
int
) – spatial dimension of the input data. Defaults to 3.in_channels (
int
) – number of input channels for the network. Default to 1.out_channels (
int
) – number of output channels for the network. Defaults to 1.psp_block_num (
int
) – the number of pyramid volumetric pooling modules used at the end of the network before the final output layer for extracting multiscale features. The number should be an integer that belongs to [0,4]. Defaults to 4.upsample_mode (
str
) –[
"transpose"
,"bilinear"
,"trilinear"
,nearest
] The mode of upsampling manipulations. Using the last two modes cannot guarantee the model’s reproducibility. Defaults totranspose
."transpose"
, uses transposed convolution layers."bilinear"
, uses bilinear interpolate."trilinear"
, uses trilinear interpolate."nearest"
, uses nearest interpolate.
pretrained (
bool
) – whether to load pretrained weights from ResNet50 to initialize convolution layers, default to False.progress (
bool
) – If True, displays a progress bar of the download of pretrained weights to stderr.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.
DenseNet¶
- 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, act=('relu', {'inplace': True}), norm='batch', dropout_prob=0.0)[source]¶
Densenet based on: Densely Connected Convolutional Networks. Adapted from PyTorch Hub 2D version: https://pytorch.org/vision/stable/models.html#id16. This network is non-determistic When spatial_dims is 3 and CUDA is enabled. Please check the link below for more details: https://pytorch.org/docs/stable/generated/torch.use_deterministic_algorithms.html#torch.use_deterministic_algorithms
- 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 (
Sequence
[int
]) – 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)act (
Union
[str
,tuple
]) – activation type and arguments. Defaults to relu.norm (
Union
[str
,tuple
]) – feature normalization type and arguments. Defaults to batch norm.dropout_prob (
float
) – dropout rate after each dense layer.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Tensor
DenseNet121¶
- class monai.networks.nets.DenseNet121(init_features=64, growth_rate=32, block_config=(6, 12, 24, 16), pretrained=False, progress=True, **kwargs)[source]¶
DenseNet121 with optional pretrained support when spatial_dims is 2.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
DenseNet169¶
- class monai.networks.nets.DenseNet169(init_features=64, growth_rate=32, block_config=(6, 12, 32, 32), pretrained=False, progress=True, **kwargs)[source]¶
DenseNet169 with optional pretrained support when spatial_dims is 2.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
DenseNet201¶
- class monai.networks.nets.DenseNet201(init_features=64, growth_rate=32, block_config=(6, 12, 48, 32), pretrained=False, progress=True, **kwargs)[source]¶
DenseNet201 with optional pretrained support when spatial_dims is 2.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
DenseNet264¶
EfficientNet¶
- class monai.networks.nets.EfficientNet(blocks_args_str, spatial_dims=2, in_channels=3, num_classes=1000, width_coefficient=1.0, depth_coefficient=1.0, dropout_rate=0.2, image_size=224, norm=('batch', {'eps': 0.001, 'momentum': 0.01}), drop_connect_rate=0.2, depth_divisor=8)[source]¶
EfficientNet based on Rethinking Model Scaling for Convolutional Neural Networks. Adapted from EfficientNet-PyTorch.
- Parameters
blocks_args_str (
List
[str
]) – block definitions.spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.num_classes (
int
) – number of output classes.width_coefficient (
float
) – width multiplier coefficient (w in paper).depth_coefficient (
float
) – depth multiplier coefficient (d in paper).dropout_rate (
float
) – dropout rate for dropout layers.image_size (
int
) – input image resolution.norm (
Union
[str
,tuple
]) – feature normalization type and arguments.drop_connect_rate (
float
) – dropconnect rate for drop connection (individual weights) layers.depth_divisor (
int
) – depth divisor for channel rounding.
- __init__(blocks_args_str, spatial_dims=2, in_channels=3, num_classes=1000, width_coefficient=1.0, depth_coefficient=1.0, dropout_rate=0.2, image_size=224, norm=('batch', {'eps': 0.001, 'momentum': 0.01}), drop_connect_rate=0.2, depth_divisor=8)[source]¶
EfficientNet based on Rethinking Model Scaling for Convolutional Neural Networks. Adapted from EfficientNet-PyTorch.
- Parameters
blocks_args_str (
List
[str
]) – block definitions.spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.num_classes (
int
) – number of output classes.width_coefficient (
float
) – width multiplier coefficient (w in paper).depth_coefficient (
float
) – depth multiplier coefficient (d in paper).dropout_rate (
float
) – dropout rate for dropout layers.image_size (
int
) – input image resolution.norm (
Union
[str
,tuple
]) – feature normalization type and arguments.drop_connect_rate (
float
) – dropconnect rate for drop connection (individual weights) layers.depth_divisor (
int
) – depth divisor for channel rounding.
BlockArgs¶
- class monai.networks.nets.BlockArgs(num_repeat: int, kernel_size: int, stride: int, expand_ratio: int, input_filters: int, output_filters: int, id_skip: bool, se_ratio: Optional[float] = None)[source]¶
- BlockArgs object to assist in decoding string notation
of arguments for MBConvBlock definition.
Create new instance of BlockArgs(num_repeat, kernel_size, stride, expand_ratio, input_filters, output_filters, id_skip, se_ratio)
- property expand_ratio¶
Alias for field number 3
- static from_string(block_string)[source]¶
Get a BlockArgs object from a string notation of arguments.
- Parameters
block_string (str) – A string notation of arguments. Examples: “r1_k3_s11_e1_i32_o16_se0.25”.
- Returns
namedtuple defined at the top of this function.
- Return type
- property id_skip¶
Alias for field number 6
- property input_filters¶
Alias for field number 4
- property kernel_size¶
Alias for field number 1
- property num_repeat¶
Alias for field number 0
- property output_filters¶
Alias for field number 5
- property se_ratio¶
Alias for field number 7
- property stride¶
Alias for field number 2
EfficientNetBN¶
- class monai.networks.nets.EfficientNetBN(model_name, pretrained=True, progress=True, spatial_dims=2, in_channels=3, num_classes=1000, norm=('batch', {'eps': 0.001, 'momentum': 0.01}), adv_prop=False)[source]¶
Generic wrapper around EfficientNet, used to initialize EfficientNet-B0 to EfficientNet-B7 models model_name is mandatory argument as there is no EfficientNetBN itself, it needs the N in [0, 1, 2, 3, 4, 5, 6, 7, 8] to be a model
- Parameters
model_name (
str
) – name of model to initialize, can be from [efficientnet-b0, …, efficientnet-b8, efficientnet-l2].pretrained (
bool
) – whether to initialize pretrained ImageNet weights, only available for spatial_dims=2 and batch norm is used.progress (
bool
) – whether to show download progress for pretrained weights download.spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.num_classes (
int
) – number of output classes.norm (
Union
[str
,tuple
]) – feature normalization type and arguments.adv_prop (
bool
) – whether to use weights trained with adversarial examples. This argument only works when pretrained is True.
Examples:
# for pretrained spatial 2D ImageNet >>> image_size = get_efficientnet_image_size("efficientnet-b0") >>> inputs = torch.rand(1, 3, image_size, image_size) >>> model = EfficientNetBN("efficientnet-b0", pretrained=True) >>> model.eval() >>> outputs = model(inputs) # create spatial 2D >>> model = EfficientNetBN("efficientnet-b0", spatial_dims=2) # create spatial 3D >>> model = EfficientNetBN("efficientnet-b0", spatial_dims=3) # create EfficientNetB7 for spatial 2D >>> model = EfficientNetBN("efficientnet-b7", spatial_dims=2)
- __init__(model_name, pretrained=True, progress=True, spatial_dims=2, in_channels=3, num_classes=1000, norm=('batch', {'eps': 0.001, 'momentum': 0.01}), adv_prop=False)[source]¶
Generic wrapper around EfficientNet, used to initialize EfficientNet-B0 to EfficientNet-B7 models model_name is mandatory argument as there is no EfficientNetBN itself, it needs the N in [0, 1, 2, 3, 4, 5, 6, 7, 8] to be a model
- Parameters
model_name (
str
) – name of model to initialize, can be from [efficientnet-b0, …, efficientnet-b8, efficientnet-l2].pretrained (
bool
) – whether to initialize pretrained ImageNet weights, only available for spatial_dims=2 and batch norm is used.progress (
bool
) – whether to show download progress for pretrained weights download.spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.num_classes (
int
) – number of output classes.norm (
Union
[str
,tuple
]) – feature normalization type and arguments.adv_prop (
bool
) – whether to use weights trained with adversarial examples. This argument only works when pretrained is True.
Examples:
# for pretrained spatial 2D ImageNet >>> image_size = get_efficientnet_image_size("efficientnet-b0") >>> inputs = torch.rand(1, 3, image_size, image_size) >>> model = EfficientNetBN("efficientnet-b0", pretrained=True) >>> model.eval() >>> outputs = model(inputs) # create spatial 2D >>> model = EfficientNetBN("efficientnet-b0", spatial_dims=2) # create spatial 3D >>> model = EfficientNetBN("efficientnet-b0", spatial_dims=3) # create EfficientNetB7 for spatial 2D >>> model = EfficientNetBN("efficientnet-b7", spatial_dims=2)
EfficientNetBNFeatures¶
- class monai.networks.nets.EfficientNetBNFeatures(model_name, pretrained=True, progress=True, spatial_dims=2, in_channels=3, num_classes=1000, norm=('batch', {'eps': 0.001, 'momentum': 0.01}), adv_prop=False)[source]¶
Initialize EfficientNet-B0 to EfficientNet-B7 models as a backbone, the backbone can be used as an encoder for segmentation and objection models. Compared with the class EfficientNetBN, the only different place is the forward function.
This class refers to PyTorch image models.
- __init__(model_name, pretrained=True, progress=True, spatial_dims=2, in_channels=3, num_classes=1000, norm=('batch', {'eps': 0.001, 'momentum': 0.01}), adv_prop=False)[source]¶
Initialize EfficientNet-B0 to EfficientNet-B7 models as a backbone, the backbone can be used as an encoder for segmentation and objection models. Compared with the class EfficientNetBN, the only different place is the forward function.
This class refers to PyTorch image models.
SegResNet¶
- class monai.networks.nets.SegResNet(spatial_dims=3, init_filters=8, in_channels=1, out_channels=2, dropout_prob=None, act=('RELU', {'inplace': True}), norm=('GROUP', {'num_groups': 8}), norm_name='', num_groups=8, use_conv_final=True, blocks_down=(1, 2, 2, 4), blocks_up=(1, 1, 1), upsample_mode=UpsampleMode.NONTRAINABLE)[source]¶
SegResNet based on 3D MRI brain tumor segmentation using autoencoder regularization. The module does not include the variational autoencoder (VAE). The model supports 2D or 3D inputs.
- Parameters
spatial_dims (
int
) – spatial dimension of the input data. Defaults to 3.init_filters (
int
) – number of output channels for initial convolution layer. Defaults to 8.in_channels (
int
) – number of input channels for the network. Defaults to 1.out_channels (
int
) – number of output channels for the network. Defaults to 2.dropout_prob (
Optional
[float
]) – probability of an element to be zero-ed. Defaults toNone
.act (
Union
[Tuple
,str
]) – activation type and arguments. Defaults toRELU
.norm (
Union
[Tuple
,str
]) – feature normalization type and arguments. Defaults toGROUP
.norm_name (
str
) – deprecating option for feature normalization type.num_groups (
int
) – deprecating option for group norm. parameters.use_conv_final (
bool
) – if add a final convolution block to output. Defaults toTrue
.blocks_down (
tuple
) – number of down sample blocks in each layer. Defaults to[1,2,2,4]
.blocks_up (
tuple
) – number of up sample blocks in each layer. Defaults to[1,1,1]
.upsample_mode (
Union
[UpsampleMode
,str
]) –[
"deconv"
,"nontrainable"
,"pixelshuffle"
] The mode of upsampling manipulations. Using thenontrainable
modes cannot guarantee the model’s reproducibility. Defaults to``nontrainable``.deconv
, uses transposed convolution layers.nontrainable
, uses non-trainable linear interpolation.pixelshuffle
, usesmonai.networks.blocks.SubpixelUpsample
.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Tensor
SegResNetVAE¶
- class monai.networks.nets.SegResNetVAE(input_image_size, vae_estimate_std=False, vae_default_std=0.3, vae_nz=256, spatial_dims=3, init_filters=8, in_channels=1, out_channels=2, dropout_prob=None, act=('RELU', {'inplace': True}), norm=('GROUP', {'num_groups': 8}), use_conv_final=True, blocks_down=(1, 2, 2, 4), blocks_up=(1, 1, 1), upsample_mode=UpsampleMode.NONTRAINABLE)[source]¶
SegResNetVAE based on 3D MRI brain tumor segmentation using autoencoder regularization. The module contains the variational autoencoder (VAE). The model supports 2D or 3D inputs.
- Parameters
input_image_size (
Sequence
[int
]) – the size of images to input into the network. It is used to determine the in_features of the fc layer in VAE.vae_estimate_std (
bool
) – whether to estimate the standard deviations in VAE. Defaults toFalse
.vae_default_std (
float
) – if not to estimate the std, use the default value. Defaults to 0.3.vae_nz (
int
) – number of latent variables in VAE. Defaults to 256. Where, 128 to represent mean, and 128 to represent std.spatial_dims (
int
) – spatial dimension of the input data. Defaults to 3.init_filters (
int
) – number of output channels for initial convolution layer. Defaults to 8.in_channels (
int
) – number of input channels for the network. Defaults to 1.out_channels (
int
) – number of output channels for the network. Defaults to 2.dropout_prob (
Optional
[float
]) – probability of an element to be zero-ed. Defaults toNone
.act (
Union
[str
,tuple
]) – activation type and arguments. Defaults toRELU
.norm (
Union
[Tuple
,str
]) – feature normalization type and arguments. Defaults toGROUP
.use_conv_final (
bool
) – if add a final convolution block to output. Defaults toTrue
.blocks_down (
tuple
) – number of down sample blocks in each layer. Defaults to[1,2,2,4]
.blocks_up (
tuple
) – number of up sample blocks in each layer. Defaults to[1,1,1]
.upsample_mode (
Union
[UpsampleMode
,str
]) –[
"deconv"
,"nontrainable"
,"pixelshuffle"
] The mode of upsampling manipulations. Using thenontrainable
modes cannot guarantee the model’s reproducibility. Defaults to``nontrainable``.deconv
, uses transposed convolution layers.nontrainable
, uses non-trainable linear interpolation.pixelshuffle
, usesmonai.networks.blocks.SubpixelUpsample
.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.
ResNet¶
- class monai.networks.nets.ResNet(block, layers, block_inplanes, spatial_dims=3, n_input_channels=3, conv1_t_size=7, conv1_t_stride=1, no_max_pool=False, shortcut_type='B', widen_factor=1.0, num_classes=400, feed_forward=True, n_classes=None)[source]¶
ResNet based on: Deep Residual Learning for Image Recognition and Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet?. Adapted from https://github.com/kenshohara/3D-ResNets-PyTorch/tree/master/models.
- Parameters
block (
Type
[Union
[ResNetBlock
,ResNetBottleneck
]]) – which ResNet block to use, either Basic or Bottleneck.layers (
List
[int
]) – how many layers to use.block_inplanes (
List
[int
]) – determine the size of planes at each step. Also tunable with widen_factor.spatial_dims (
int
) – number of spatial dimensions of the input image.n_input_channels (
int
) – number of input channels for first convolutional layer.conv1_t_size (
int
) – size of first convolution layer, determines kernel and padding.conv1_t_stride (
int
) – stride of first convolution layer.no_max_pool (
bool
) – bool argument to determine if to use maxpool layer.shortcut_type (
str
) – which downsample block to use. Options are ‘A’, ‘B’, default to ‘B’. - ‘A’: using self._downsample_basic_block. - ‘B’: kernel_size 1 conv + norm.widen_factor (
float
) – widen output for each layer.num_classes (
int
) – number of output (classifications).feed_forward (
bool
) – whether to add the FC layer for the output, default to True.
Deprecated since version 0.6.0:
n_classes
is deprecated, usenum_classes
instead.Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Tensor
SENet¶
- class monai.networks.nets.SENet(spatial_dims, in_channels, block, layers, groups, reduction, dropout_prob=0.2, dropout_dim=1, inplanes=128, downsample_kernel_size=3, input_3x3=True, num_classes=1000)[source]¶
SENet based on Squeeze-and-Excitation Networks. Adapted from Cadene Hub 2D version.
- Parameters
spatial_dims (
int
) – spatial dimension of the input data.in_channels (
int
) – channel number of the input data.block (
Type
[Union
[SEBottleneck
,SEResNetBottleneck
,SEResNeXtBottleneck
]]) – SEBlock class. for SENet154: SEBottleneck for SE-ResNet models: SEResNetBottleneck for SE-ResNeXt models: SEResNeXtBottlenecklayers (
Sequence
[int
]) – number of residual blocks for 4 layers of the network (layer1…layer4).groups (
int
) – number of groups for the 3x3 convolution in each bottleneck block. for SENet154: 64 for SE-ResNet models: 1 for SE-ResNeXt models: 32reduction (
int
) – reduction ratio for Squeeze-and-Excitation modules. for all models: 16dropout_prob (
Optional
[float
]) – drop probability for the Dropout layer. if None the Dropout layer is not used. for SENet154: 0.2 for SE-ResNet models: None for SE-ResNeXt models: Nonedropout_dim (
int
) – determine the dimensions of dropout. Defaults to 1. When dropout_dim = 1, randomly zeroes some of the elements for each channel. When dropout_dim = 2, Randomly zeroes out entire channels (a channel is a 2D feature map). When dropout_dim = 3, Randomly zeroes out entire channels (a channel is a 3D feature map).inplanes (
int
) – number of input channels for layer1. for SENet154: 128 for SE-ResNet models: 64 for SE-ResNeXt models: 64downsample_kernel_size (
int
) – kernel size for downsampling convolutions in layer2, layer3 and layer4. for SENet154: 3 for SE-ResNet models: 1 for SE-ResNeXt models: 1input_3x3 (
bool
) – If True, use three 3x3 convolutions instead of a single 7x7 convolution in layer0. - For SENet154: True - For SE-ResNet models: False - For SE-ResNeXt models: Falsenum_classes (
int
) – number of outputs in last_linear layer. for all models: 1000
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Tensor
SENet154¶
- class monai.networks.nets.SENet154(layers=(3, 8, 36, 3), groups=64, reduction=16, pretrained=False, progress=True, **kwargs)[source]¶
SENet154 based on Squeeze-and-Excitation Networks with optional pretrained support when spatial_dims is 2.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
SEResNet50¶
- class monai.networks.nets.SEResNet50(layers=(3, 4, 6, 3), groups=1, reduction=16, dropout_prob=None, inplanes=64, downsample_kernel_size=1, input_3x3=False, pretrained=False, progress=True, **kwargs)[source]¶
SEResNet50 based on Squeeze-and-Excitation Networks with optional pretrained support when spatial_dims is 2.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
SEResNet101¶
- class monai.networks.nets.SEResNet101(layers=(3, 4, 23, 3), groups=1, reduction=16, inplanes=64, downsample_kernel_size=1, input_3x3=False, pretrained=False, progress=True, **kwargs)[source]¶
SEResNet101 based on Squeeze-and-Excitation Networks with optional pretrained support when spatial_dims is 2.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
SEResNet152¶
- class monai.networks.nets.SEResNet152(layers=(3, 8, 36, 3), groups=1, reduction=16, inplanes=64, downsample_kernel_size=1, input_3x3=False, pretrained=False, progress=True, **kwargs)[source]¶
SEResNet152 based on Squeeze-and-Excitation Networks with optional pretrained support when spatial_dims is 2.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
SEResNext50¶
- class monai.networks.nets.SEResNext50(layers=(3, 4, 6, 3), groups=32, reduction=16, dropout_prob=None, inplanes=64, downsample_kernel_size=1, input_3x3=False, pretrained=False, progress=True, **kwargs)[source]¶
SEResNext50 based on Squeeze-and-Excitation Networks with optional pretrained support when spatial_dims is 2.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
SEResNext101¶
- class monai.networks.nets.SEResNext101(layers=(3, 4, 23, 3), groups=32, reduction=16, dropout_prob=None, inplanes=64, downsample_kernel_size=1, input_3x3=False, pretrained=False, progress=True, **kwargs)[source]¶
SEResNext101 based on Squeeze-and-Excitation Networks with optional pretrained support when spatial_dims is 2.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
HighResNet¶
- class monai.networks.nets.HighResNet(spatial_dims=3, in_channels=1, out_channels=1, norm_type=('batch', {'affine': True}), acti_type=('relu', {'inplace': True}), dropout_prob=0.0, bias=False, 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}), channel_matching=ChannelMatching.PAD)[source]¶
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
[str
,tuple
]) – feature normalization type and arguments. Defaults to("batch", {"affine": True})
.acti_type (
Union
[str
,tuple
]) – activation type and arguments. Defaults to("relu", {"inplace": True})
.dropout_prob (
Union
[Tuple
,str
,float
,None
]) – probability of the feature map to be zeroed (only applies to the penultimate conv layer).bias (
bool
) –whether to have a bias term in convolution blocks. Defaults to False. According to Performance Tuning Guide, if a conv layer is directly followed by a batch norm layer, bias should be False.
layer_params (
Sequence
[Dict
]) – specifying key parameters of each layer/block.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 one.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Tensor
- class monai.networks.nets.HighResBlock(spatial_dims, in_channels, out_channels, kernels=(3, 3), dilation=1, norm_type=('batch', {'affine': True}), acti_type=('relu', {'inplace': True}), bias=False, channel_matching=ChannelMatching.PAD)[source]¶
- 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.kernels (
Sequence
[int
]) – each integer k in kernels corresponds to a convolution layer with kernel size k.dilation (
Union
[Sequence
[int
],int
]) – spacing between kernel elements.norm_type (
Union
[Tuple
,str
]) – feature normalization type and arguments. Defaults to("batch", {"affine": True})
.acti_type (
Union
[Tuple
,str
]) – {"relu"
,"prelu"
,"relu6"
} Non-linear activation using ReLU or PReLU. Defaults to"relu"
.bias (
bool
) –whether to have a bias term in convolution blocks. Defaults to False. According to Performance Tuning Guide, if a conv layer is directly followed by a batch norm layer, bias should be False.
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 one.
- Raises
ValueError – When
channel_matching=pad
andin_channels > out_channels
. Incompatible values.
- __init__(spatial_dims, in_channels, out_channels, kernels=(3, 3), dilation=1, norm_type=('batch', {'affine': True}), acti_type=('relu', {'inplace': True}), bias=False, channel_matching=ChannelMatching.PAD)[source]¶
- 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.kernels (
Sequence
[int
]) – each integer k in kernels corresponds to a convolution layer with kernel size k.dilation (
Union
[Sequence
[int
],int
]) – spacing between kernel elements.norm_type (
Union
[Tuple
,str
]) – feature normalization type and arguments. Defaults to("batch", {"affine": True})
.acti_type (
Union
[Tuple
,str
]) – {"relu"
,"prelu"
,"relu6"
} Non-linear activation using ReLU or PReLU. Defaults to"relu"
.bias (
bool
) –whether to have a bias term in convolution blocks. Defaults to False. According to Performance Tuning Guide, if a conv layer is directly followed by a batch norm layer, bias should be False.
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 one.
- Raises
ValueError – When
channel_matching=pad
andin_channels > out_channels
. Incompatible values.
- forward(x)[source]¶
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.- Return type
Tensor
DynUNet¶
- class monai.networks.nets.DynUNet(spatial_dims, in_channels, out_channels, kernel_size, strides, upsample_kernel_size, filters=None, dropout=None, norm_name=('INSTANCE', {'affine': True}), act_name=('leakyrelu', {'inplace': True, 'negative_slope': 0.01}), deep_supervision=False, deep_supr_num=1, res_block=False, trans_bias=False)[source]¶
This reimplementation of a dynamic UNet (DynUNet) is based on: Automated Design of Deep Learning Methods for Biomedical Image Segmentation. nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation. Optimized U-Net for Brain Tumor Segmentation.
This model is more flexible compared with
monai.networks.nets.UNet
in three places:Residual connection is supported in conv blocks.
Anisotropic kernel sizes and strides can be used in each layers.
Deep supervision heads can be added.
The model supports 2D or 3D inputs and is consisted with four kinds of blocks: one input block, n downsample blocks, one bottleneck and n+1 upsample blocks. Where, n>0. The first and last kernel and stride values of the input sequences are used for input block and bottleneck respectively, and the rest value(s) are used for downsample and upsample blocks. Therefore, pleasure ensure that the length of input sequences (
kernel_size
andstrides
) is no less than 3 in order to have at least one downsample and upsample blocks.To meet the requirements of the structure, the input size for each spatial dimension should be divisible by 2 * the product of all strides in the corresponding dimension. The output size for each spatial dimension equals to the input size of the corresponding dimension divided by the stride in strides[0]. For example, if strides=((1, 2, 4), 2, 1, 1), the minimal spatial size of the input is (8, 16, 32), and the spatial size of the output is (8, 8, 8).
For backwards compatibility with old weights, please set strict=False when calling load_state_dict.
Usage example with medical segmentation decathlon dataset is available at: https://github.com/Project-MONAI/tutorials/tree/master/modules/dynunet_pipeline.
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.kernel_size (
Sequence
[Union
[Sequence
[int
],int
]]) – convolution kernel size.strides (
Sequence
[Union
[Sequence
[int
],int
]]) – convolution strides for each blocks.upsample_kernel_size (
Sequence
[Union
[Sequence
[int
],int
]]) – convolution kernel size for transposed convolution layers. The values should equal to strides[1:].filters (
Optional
[Sequence
[int
]]) – number of output channels for each blocks. Different from nnU-Net, in this implementation we add this argument to make the network more flexible. As shown in the third reference, one way to determine this argument is like:[64, 96, 128, 192, 256, 384, 512, 768, 1024][: len(strides)]
. The above way is used in the network that wins task 1 in the BraTS21 Challenge. If not specified, the way which nnUNet used will be employed. Defaults toNone
.dropout (
Union
[Tuple
,str
,float
,None
]) – dropout ratio. Defaults to no dropout.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments. Defaults toINSTANCE
.act_name (
Union
[Tuple
,str
]) – activation layer type and arguments. Defaults toleakyrelu
.deep_supervision (
bool
) – whether to add deep supervision head before output. Defaults toFalse
. IfTrue
, in training mode, the forward function will output not only the final feature map (from output_block), but also the feature maps that come from the intermediate up sample layers. In order to unify the return type (the restriction of TorchScript), all intermediate feature maps are interpolated into the same size as the final feature map and stacked together (with a new dimension in the first axis)into one single tensor. For instance, if there are two intermediate feature maps with shapes: (1, 2, 16, 12) and (1, 2, 8, 6), and the final feature map has the shape (1, 2, 32, 24), then all intermediate feature maps will be interpolated into (1, 2, 32, 24), and the stacked tensor will has the shape (1, 3, 2, 32, 24). When calculating the loss, you can use torch.unbind to get all feature maps can compute the loss one by one with the ground truth, then do a weighted average for all losses to achieve the final loss.deep_supr_num (
int
) – number of feature maps that will output during deep supervision head. The value should be larger than 0 and less than the number of up sample layers. Defaults to 1.res_block (
bool
) – whether to use residual connection based convolution blocks during the network. Defaults toFalse
.trans_bias (
bool
) – whether to set the bias parameter in transposed convolution layers. Defaults toFalse
.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.DynUnet¶
alias of
monai.networks.nets.dynunet.DynUNet
- monai.networks.nets.Dynunet¶
alias of
monai.networks.nets.dynunet.DynUNet
UNet¶
- class monai.networks.nets.UNet(spatial_dims, in_channels, out_channels, channels, strides, kernel_size=3, up_kernel_size=3, num_res_units=0, act='PRELU', norm='INSTANCE', dropout=0.0, bias=True, dimensions=None)[source]¶
Enhanced version of UNet which has residual units implemented with the ResidualUnit class. The residual part uses a convolution to change the input dimensions to match the output dimensions if this is necessary but will use nn.Identity if not. Refer to: https://link.springer.com/chapter/10.1007/978-3-030-12029-0_40.
Each layer of the network has a encode and decode path with a skip connection between them. Data in the encode path is downsampled using strided convolutions (if strides is given values greater than 1) and in the decode path upsampled using strided transpose convolutions. These down or up sampling operations occur at the beginning of each block rather than afterwards as is typical in UNet implementations.
To further explain this consider the first example network given below. This network has 3 layers with strides of 2 for each of the middle layers (the last layer is the bottom connection which does not down/up sample). Input data to this network is immediately reduced in the spatial dimensions by a factor of 2 by the first convolution of the residual unit defining the first layer of the encode part. The last layer of the decode part will upsample its input (data from the previous layer concatenated with data from the skip connection) in the first convolution. this ensures the final output of the network has the same shape as the input.
Padding values for the convolutions are chosen to ensure output sizes are even divisors/multiples of the input sizes if the strides value for a layer is a factor of the input sizes. A typical case is to use strides values of 2 and inputs that are multiples of powers of 2. An input can thus be downsampled evenly however many times its dimensions can be divided by 2, so for the example network inputs would have to have dimensions that are multiples of 4. In the second example network given below the input to the bottom layer will have shape (1, 64, 15, 15) for an input of shape (1, 1, 240, 240) demonstrating the input being reduced in size spatially by 2**4.
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.channels (
Sequence
[int
]) – sequence of channels. Top block first. The length of channels should be no less than 2.strides (
Sequence
[int
]) – sequence of convolution strides. The length of stride should equal to len(channels) - 1.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size, the value(s) should be odd. If sequence, its length should equal to dimensions. Defaults to 3.up_kernel_size (
Union
[Sequence
[int
],int
]) – upsampling convolution kernel size, the value(s) should be odd. If sequence, its length should equal to dimensions. Defaults to 3.num_res_units (
int
) – number of residual units. Defaults to 0.act (
Union
[Tuple
,str
]) – activation type and arguments. Defaults to PReLU.norm (
Union
[Tuple
,str
]) – feature normalization type and arguments. Defaults to instance norm.dropout (
float
) – dropout ratio. Defaults to no dropout.bias (
bool
) –whether to have a bias term in convolution blocks. Defaults to True. According to Performance Tuning Guide, if a conv layer is directly followed by a batch norm layer, bias should be False.
Examples:
from monai.networks.nets import UNet # 3 layer network with down/upsampling by a factor of 2 at each layer with 2-convolution residual units net = UNet( spatial_dims=2, in_channels=1, out_channels=1, channels=(4, 8, 16), strides=(2, 2), num_res_units=2 ) # 5 layer network with simple convolution/normalization/dropout/activation blocks defining the layers net=UNet( spatial_dims=2, in_channels=1, out_channels=1, channels=(4, 8, 16, 32, 64), strides=(2, 2, 2, 2), )
Deprecated since version 0.6.0:
dimensions
is deprecated, usespatial_dims
instead.- Note: The acceptable spatial size of input data depends on the parameters of the network,
to set appropriate spatial size, please check the tutorial for more details: https://github.com/Project-MONAI/tutorials/blob/master/modules/UNet_input_size_constrains.ipynb. Typically, when using a stride of 2 in down / up sampling, the output dimensions are either half of the input when downsampling, or twice when upsampling. In this case with N numbers of layers in the network, the inputs must have spatial dimensions that are all multiples of 2^N. Usually, applying resize, pad or crop transforms can help adjust the spatial size of input data.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Tensor
- monai.networks.nets.Unet¶
alias of
monai.networks.nets.unet.UNet
- monai.networks.nets.unet¶
alias of <module ‘monai.networks.nets.unet’ from ‘/home/docs/checkouts/readthedocs.org/user_builds/monai/checkouts/stable/monai/networks/nets/unet.py’>
UNETR¶
- class monai.networks.nets.UNETR(in_channels, out_channels, img_size, feature_size=16, hidden_size=768, mlp_dim=3072, num_heads=12, pos_embed='conv', norm_name='instance', conv_block=True, res_block=True, dropout_rate=0.0, spatial_dims=3)[source]¶
UNETR based on: “Hatamizadeh et al., UNETR: Transformers for 3D Medical Image Segmentation <https://arxiv.org/abs/2103.10504>”
- Parameters
in_channels (
int
) – dimension of input channels.out_channels (
int
) – dimension of output channels.img_size (
Union
[Sequence
[int
],int
]) – dimension of input image.feature_size (
int
) – dimension of network feature size.hidden_size (
int
) – dimension of hidden layer.mlp_dim (
int
) – dimension of feedforward layer.num_heads (
int
) – number of attention heads.pos_embed (
str
) – position embedding layer type.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.conv_block (
bool
) – bool argument to determine if convolutional block is used.res_block (
bool
) – bool argument to determine if residual block is used.dropout_rate (
float
) – faction of the input units to drop.spatial_dims (
int
) – number of spatial dims.
Examples:
# for single channel input 4-channel output with image size of (96,96,96), feature size of 32 and batch norm >>> net = UNETR(in_channels=1, out_channels=4, img_size=(96,96,96), feature_size=32, norm_name='batch') # for single channel input 4-channel output with image size of (96,96), feature size of 32 and batch norm >>> net = UNETR(in_channels=1, out_channels=4, img_size=96, feature_size=32, norm_name='batch', spatial_dims=2) # for 4-channel input 3-channel output with image size of (128,128,128), conv position embedding and instance norm >>> net = UNETR(in_channels=4, out_channels=3, img_size=(128,128,128), pos_embed='conv', norm_name='instance')
- __init__(in_channels, out_channels, img_size, feature_size=16, hidden_size=768, mlp_dim=3072, num_heads=12, pos_embed='conv', norm_name='instance', conv_block=True, res_block=True, dropout_rate=0.0, spatial_dims=3)[source]¶
- Parameters
in_channels (
int
) – dimension of input channels.out_channels (
int
) – dimension of output channels.img_size (
Union
[Sequence
[int
],int
]) – dimension of input image.feature_size (
int
) – dimension of network feature size.hidden_size (
int
) – dimension of hidden layer.mlp_dim (
int
) – dimension of feedforward layer.num_heads (
int
) – number of attention heads.pos_embed (
str
) – position embedding layer type.norm_name (
Union
[Tuple
,str
]) – feature normalization type and arguments.conv_block (
bool
) – bool argument to determine if convolutional block is used.res_block (
bool
) – bool argument to determine if residual block is used.dropout_rate (
float
) – faction of the input units to drop.spatial_dims (
int
) – number of spatial dims.
Examples:
# for single channel input 4-channel output with image size of (96,96,96), feature size of 32 and batch norm >>> net = UNETR(in_channels=1, out_channels=4, img_size=(96,96,96), feature_size=32, norm_name='batch') # for single channel input 4-channel output with image size of (96,96), feature size of 32 and batch norm >>> net = UNETR(in_channels=1, out_channels=4, img_size=96, feature_size=32, norm_name='batch', spatial_dims=2) # for 4-channel input 3-channel output with image size of (128,128,128), conv position embedding and instance norm >>> net = UNETR(in_channels=4, out_channels=3, img_size=(128,128,128), pos_embed='conv', norm_name='instance')
- forward(x_in)[source]¶
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.
BasicUNet¶
- class monai.networks.nets.BasicUNet(spatial_dims=3, in_channels=1, out_channels=2, features=(32, 32, 64, 128, 256, 32), act=('LeakyReLU', {'negative_slope': 0.1, 'inplace': True}), norm=('instance', {'affine': True}), bias=True, dropout=0.0, upsample='deconv', dimensions=None)[source]¶
A UNet implementation with 1D/2D/3D supports.
Based on:
Falk et al. “U-Net – Deep Learning for Cell Counting, Detection, and Morphometry”. Nature Methods 16, 67–70 (2019), DOI: http://dx.doi.org/10.1038/s41592-018-0261-2
- Parameters
spatial_dims (
int
) – number of spatial dimensions. Defaults to 3 for spatial 3D inputs.in_channels (
int
) – number of input channels. Defaults to 1.out_channels (
int
) – number of output channels. Defaults to 2.features (
Sequence
[int
]) –six integers as numbers of features. Defaults to
(32, 32, 64, 128, 256, 32)
,the first five values correspond to the five-level encoder feature sizes.
the last value corresponds to the feature size after the last upsampling.
act (
Union
[str
,tuple
]) – activation type and arguments. Defaults to LeakyReLU.norm (
Union
[str
,tuple
]) – feature normalization type and arguments. Defaults to instance norm.bias (
bool
) –whether to have a bias term in convolution blocks. Defaults to True. According to Performance Tuning Guide, if a conv layer is directly followed by a batch norm layer, bias should be False.
dropout (
Union
[float
,tuple
]) – dropout ratio. Defaults to no dropout.upsample (
str
) – upsampling mode, available options are"deconv"
,"pixelshuffle"
,"nontrainable"
.
Deprecated since version 0.6.0:
dimensions
is deprecated, usespatial_dims
instead.Examples:
# for spatial 2D >>> net = BasicUNet(spatial_dims=2, features=(64, 128, 256, 512, 1024, 128)) # for spatial 2D, with group norm >>> net = BasicUNet(spatial_dims=2, features=(64, 128, 256, 512, 1024, 128), norm=("group", {"num_groups": 4})) # for spatial 3D >>> net = BasicUNet(spatial_dims=3, features=(32, 32, 64, 128, 256, 32))
See Also
- __init__(spatial_dims=3, in_channels=1, out_channels=2, features=(32, 32, 64, 128, 256, 32), act=('LeakyReLU', {'negative_slope': 0.1, 'inplace': True}), norm=('instance', {'affine': True}), bias=True, dropout=0.0, upsample='deconv', dimensions=None)[source]¶
A UNet implementation with 1D/2D/3D supports.
Based on:
Falk et al. “U-Net – Deep Learning for Cell Counting, Detection, and Morphometry”. Nature Methods 16, 67–70 (2019), DOI: http://dx.doi.org/10.1038/s41592-018-0261-2
- Parameters
spatial_dims (
int
) – number of spatial dimensions. Defaults to 3 for spatial 3D inputs.in_channels (
int
) – number of input channels. Defaults to 1.out_channels (
int
) – number of output channels. Defaults to 2.features (
Sequence
[int
]) –six integers as numbers of features. Defaults to
(32, 32, 64, 128, 256, 32)
,the first five values correspond to the five-level encoder feature sizes.
the last value corresponds to the feature size after the last upsampling.
act (
Union
[str
,tuple
]) – activation type and arguments. Defaults to LeakyReLU.norm (
Union
[str
,tuple
]) – feature normalization type and arguments. Defaults to instance norm.bias (
bool
) –whether to have a bias term in convolution blocks. Defaults to True. According to Performance Tuning Guide, if a conv layer is directly followed by a batch norm layer, bias should be False.
dropout (
Union
[float
,tuple
]) – dropout ratio. Defaults to no dropout.upsample (
str
) – upsampling mode, available options are"deconv"
,"pixelshuffle"
,"nontrainable"
.
Deprecated since version 0.6.0:
dimensions
is deprecated, usespatial_dims
instead.Examples:
# for spatial 2D >>> net = BasicUNet(spatial_dims=2, features=(64, 128, 256, 512, 1024, 128)) # for spatial 2D, with group norm >>> net = BasicUNet(spatial_dims=2, features=(64, 128, 256, 512, 1024, 128), norm=("group", {"num_groups": 4})) # for spatial 3D >>> net = BasicUNet(spatial_dims=3, features=(32, 32, 64, 128, 256, 32))
See Also
- forward(x)[source]¶
- Parameters
x (
Tensor
) – input should have spatially N dimensions(Batch, in_channels, dim_0[, dim_1, ..., dim_N])
, N is defined by dimensions. It is recommended to havedim_n % 16 == 0
to ensure all maxpooling inputs have even edge lengths.- Returns
A torch Tensor of “raw” predictions in shape
(Batch, out_channels, dim_0[, dim_1, ..., dim_N])
.
- monai.networks.nets.BasicUnet¶
- monai.networks.nets.Basicunet¶
VNet¶
- class monai.networks.nets.VNet(spatial_dims=3, in_channels=1, out_channels=1, act=('elu', {'inplace': True}), dropout_prob=0.5, dropout_dim=3, bias=False)[source]¶
V-Net based on Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation. Adapted from the official Caffe implementation. and another pytorch implementation. The model supports 2D or 3D inputs.
- Parameters
spatial_dims (
int
) – spatial dimension of the input data. Defaults to 3.in_channels (
int
) – number of input channels for the network. Defaults to 1. The value should meet the condition that16 % in_channels == 0
.out_channels (
int
) – number of output channels for the network. Defaults to 1.act (
Union
[Tuple
[str
,Dict
],str
]) – activation type in the network. Defaults to("elu", {"inplace": True})
.dropout_prob (
float
) – dropout ratio. Defaults to 0.5. Defaults to 3.dropout_dim (
int
) –determine the dimensions of dropout. Defaults to 3.
dropout_dim = 1
, randomly zeroes some of the elements for each channel.dropout_dim = 2
, Randomly zeroes out entire channels (a channel is a 2D feature map).dropout_dim = 3
, Randomly zeroes out entire channels (a channel is a 3D feature map).
bias (
bool
) –whether to have a bias term in convolution blocks. Defaults to False. According to Performance Tuning Guide, if a conv layer is directly followed by a batch norm layer, bias should be False.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.
RegUNet¶
- class monai.networks.nets.RegUNet(spatial_dims, in_channels, num_channel_initial, depth, out_kernel_initializer='kaiming_uniform', out_activation=None, out_channels=3, extract_levels=None, pooling=True, concat_skip=False, encode_kernel_sizes=3)[source]¶
Class that implements an adapted UNet. This class also serve as the parent class of LocalNet and GlobalNet
- Reference:
O. Ronneberger, P. Fischer, and T. Brox, “U-net: Convolutional networks for biomedical image segmentation,”, Lecture Notes in Computer Science, 2015, vol. 9351, pp. 234–241. https://arxiv.org/abs/1505.04597
- Adapted from:
DeepReg (https://github.com/DeepRegNet/DeepReg)
- Parameters
spatial_dims (
int
) – number of spatial dimsin_channels (
int
) – number of input channelsnum_channel_initial (
int
) – number of initial channelsdepth (
int
) – input is at level 0, bottom is at level depth.out_kernel_initializer (
Optional
[str
]) – kernel initializer for the last layerout_activation (
Optional
[str
]) – activation at the last layerout_channels (
int
) – number of channels for the outputextract_levels (
Optional
[Tuple
[int
]]) – list, which levels from net to extract. The maximum level must equal todepth
pooling (
bool
) – for down-sampling, use non-parameterized pooling if true, otherwise use conv3dconcat_skip (
bool
) – when up-sampling, concatenate skipped tensor if true, otherwise use additionencode_kernel_sizes (
Union
[int
,List
[int
]]) – kernel size for down-sampling
- __init__(spatial_dims, in_channels, num_channel_initial, depth, out_kernel_initializer='kaiming_uniform', out_activation=None, out_channels=3, extract_levels=None, pooling=True, concat_skip=False, encode_kernel_sizes=3)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimsin_channels (
int
) – number of input channelsnum_channel_initial (
int
) – number of initial channelsdepth (
int
) – input is at level 0, bottom is at level depth.out_kernel_initializer (
Optional
[str
]) – kernel initializer for the last layerout_activation (
Optional
[str
]) – activation at the last layerout_channels (
int
) – number of channels for the outputextract_levels (
Optional
[Tuple
[int
]]) – list, which levels from net to extract. The maximum level must equal todepth
pooling (
bool
) – for down-sampling, use non-parameterized pooling if true, otherwise use conv3dconcat_skip (
bool
) – when up-sampling, concatenate skipped tensor if true, otherwise use additionencode_kernel_sizes (
Union
[int
,List
[int
]]) – kernel size for down-sampling
GlobalNet¶
- class monai.networks.nets.GlobalNet(image_size, spatial_dims, in_channels, num_channel_initial, depth, out_kernel_initializer='kaiming_uniform', out_activation=None, pooling=True, concat_skip=False, encode_kernel_sizes=3)[source]¶
Build GlobalNet for image registration.
- Reference:
Hu, Yipeng, et al. “Label-driven weakly-supervised learning for multimodal deformable image registration,” https://arxiv.org/abs/1711.01666
- Parameters
spatial_dims (
int
) – number of spatial dimsin_channels (
int
) – number of input channelsnum_channel_initial (
int
) – number of initial channelsdepth (
int
) – input is at level 0, bottom is at level depth.out_kernel_initializer (
Optional
[str
]) – kernel initializer for the last layerout_activation (
Optional
[str
]) – activation at the last layerout_channels – number of channels for the output
extract_levels – list, which levels from net to extract. The maximum level must equal to
depth
pooling (
bool
) – for down-sampling, use non-parameterized pooling if true, otherwise use conv3dconcat_skip (
bool
) – when up-sampling, concatenate skipped tensor if true, otherwise use additionencode_kernel_sizes (
Union
[int
,List
[int
]]) – kernel size for down-sampling
LocalNet¶
- class monai.networks.nets.LocalNet(spatial_dims, in_channels, num_channel_initial, extract_levels, out_kernel_initializer='kaiming_uniform', out_activation=None, out_channels=3, pooling=True, concat_skip=False)[source]¶
Reimplementation of LocalNet, based on: Weakly-supervised convolutional neural networks for multimodal image registration. Label-driven weakly-supervised learning for multimodal deformable image registration.
- Adapted from:
DeepReg (https://github.com/DeepRegNet/DeepReg)
- Parameters
spatial_dims (
int
) – number of spatial dimsin_channels (
int
) – number of input channelsnum_channel_initial (
int
) – number of initial channelsout_kernel_initializer (
Optional
[str
]) – kernel initializer for the last layerout_activation (
Optional
[str
]) – activation at the last layerout_channels (
int
) – number of channels for the outputextract_levels (
Tuple
[int
]) – list, which levels from net to extract. The maximum level must equal todepth
pooling (
bool
) – for down-sampling, use non-parameterized pooling if true, otherwise use conv3dconcat_skip (
bool
) – when up-sampling, concatenate skipped tensor if true, otherwise use addition
- __init__(spatial_dims, in_channels, num_channel_initial, extract_levels, out_kernel_initializer='kaiming_uniform', out_activation=None, out_channels=3, pooling=True, concat_skip=False)[source]¶
- Parameters
spatial_dims (
int
) – number of spatial dimsin_channels (
int
) – number of input channelsnum_channel_initial (
int
) – number of initial channelsout_kernel_initializer (
Optional
[str
]) – kernel initializer for the last layerout_activation (
Optional
[str
]) – activation at the last layerout_channels (
int
) – number of channels for the outputextract_levels (
Tuple
[int
]) – list, which levels from net to extract. The maximum level must equal todepth
pooling (
bool
) – for down-sampling, use non-parameterized pooling if true, otherwise use conv3dconcat_skip (
bool
) – when up-sampling, concatenate skipped tensor if true, otherwise use addition
AutoEncoder¶
- class monai.networks.nets.AutoEncoder(spatial_dims, in_channels, out_channels, channels, strides, kernel_size=3, up_kernel_size=3, num_res_units=0, inter_channels=None, inter_dilations=None, num_inter_units=2, act='PRELU', norm='INSTANCE', dropout=None, bias=True, dimensions=None)[source]¶
Simple definition of an autoencoder and base class for the architecture implementing
monai.networks.nets.VarAutoEncoder
. The network is composed of an encode sequence of blocks, followed by an intermediary sequence of blocks, and finally a decode sequence of blocks. The encode and decode blocks are defaultmonai.networks.blocks.Convolution
instances with the encode blocks having the given stride and the decode blocks having transpose convolutions with the same stride. If num_res_units is given residual blocks are used instead.By default the intermediary sequence is empty but if inter_channels is given to specify the output channels of blocks then this will be become a sequence of Convolution blocks or of residual blocks if num_inter_units is given. The optional parameter inter_dilations can be used to specify the dilation values of the convolutions in these blocks, this allows a network to use dilated kernels in this middle section. Since the intermediary section isn’t meant to change the size of the output the strides for all these kernels is 1.
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.channels (
Sequence
[int
]) – sequence of channels. Top block first. The length of channels should be no less than 2.strides (
Sequence
[int
]) – sequence of convolution strides. The length of stride should equal to len(channels) - 1.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size, the value(s) should be odd. If sequence, its length should equal to dimensions. Defaults to 3.up_kernel_size (
Union
[Sequence
[int
],int
]) – upsampling convolution kernel size, the value(s) should be odd. If sequence, its length should equal to dimensions. Defaults to 3.num_res_units (
int
) – number of residual units. Defaults to 0.inter_channels (
Optional
[list
]) – sequence of channels defining the blocks in the intermediate layer between encode and decode.inter_dilations (
Optional
[list
]) – defines the dilation value for each block of the intermediate layer. Defaults to 1.num_inter_units (
int
) – number of residual units for each block of the intermediate layer. Defaults to 0.act (
Union
[Tuple
,str
,None
]) – activation type and arguments. Defaults to PReLU.norm (
Union
[Tuple
,str
]) – feature normalization type and arguments. Defaults to instance norm.dropout (
Union
[Tuple
,str
,float
,None
]) – dropout ratio. Defaults to no dropout.bias (
bool
) –whether to have a bias term in convolution blocks. Defaults to True. According to Performance Tuning Guide, if a conv layer is directly followed by a batch norm layer, bias should be False.
Deprecated since version 0.6.0:
dimensions
is deprecated, usespatial_dims
instead.Examples:
from monai.networks.nets import AutoEncoder # 3 layers each down/up sampling their inputs by a factor 2 with no intermediate layer net = AutoEncoder( spatial_dims=2, in_channels=1, out_channels=1, channels=(2, 4, 8), strides=(2, 2, 2) ) # 1 layer downsampling by 2, followed by a sequence of residual units with 2 convolutions defined by # progressively increasing dilations, then final upsample layer net = AutoEncoder( spatial_dims=2, in_channels=1, out_channels=1, channels=(4,), strides=(2,), inter_channels=(8, 8, 8), inter_dilations=(1, 2, 4), num_inter_units=2 )
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Any
VarAutoEncoder¶
- class monai.networks.nets.VarAutoEncoder(spatial_dims, in_shape, out_channels, latent_size, channels, strides, kernel_size=3, up_kernel_size=3, num_res_units=0, inter_channels=None, inter_dilations=None, num_inter_units=2, act='PRELU', norm='INSTANCE', dropout=None, bias=True, dimensions=None)[source]¶
Variational Autoencoder based on the paper - https://arxiv.org/abs/1312.6114
- Parameters
spatial_dims (
int
) – number of spatial dimensions.in_shape (
Sequence
[int
]) – shape of input data starting with channel dimension.out_channels (
int
) – number of output channels.latent_size (
int
) – size of the latent variable.channels (
Sequence
[int
]) – sequence of channels. Top block first. The length of channels should be no less than 2.strides (
Sequence
[int
]) – sequence of convolution strides. The length of stride should equal to len(channels) - 1.kernel_size (
Union
[Sequence
[int
],int
]) – convolution kernel size, the value(s) should be odd. If sequence, its length should equal to dimensions. Defaults to 3.up_kernel_size (
Union
[Sequence
[int
],int
]) – upsampling convolution kernel size, the value(s) should be odd. If sequence, its length should equal to dimensions. Defaults to 3.num_res_units (
int
) – number of residual units. Defaults to 0.inter_channels (
Optional
[list
]) – sequence of channels defining the blocks in the intermediate layer between encode and decode.inter_dilations (
Optional
[list
]) – defines the dilation value for each block of the intermediate layer. Defaults to 1.num_inter_units (
int
) – number of residual units for each block of the intermediate layer. Defaults to 0.act (
Union
[Tuple
,str
,None
]) – activation type and arguments. Defaults to PReLU.norm (
Union
[Tuple
,str
]) – feature normalization type and arguments. Defaults to instance norm.dropout (
Union
[Tuple
,str
,float
,None
]) – dropout ratio. Defaults to no dropout.bias (
bool
) –whether to have a bias term in convolution blocks. Defaults to True. According to Performance Tuning Guide, if a conv layer is directly followed by a batch norm layer, bias should be False.
Deprecated since version 0.6.0:
dimensions
is deprecated, usespatial_dims
instead.Examples:
from monai.networks.nets import VarAutoEncoder # 3 layer network accepting images with dimensions (1, 32, 32) and using a latent vector with 2 values model = VarAutoEncoder( dimensions=2, in_shape=(32, 32), # image spatial shape out_channels=1, latent_size=2, channels=(16, 32, 64), strides=(1, 2, 2), )
See also
Variational autoencoder network with MedNIST Dataset https://github.com/Project-MONAI/tutorials/blob/master/modules/varautoencoder_mednist.ipynb
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Tuple
[Tensor
,Tensor
,Tensor
,Tensor
]
ViT¶
- class monai.networks.nets.ViT(in_channels, img_size, patch_size, hidden_size=768, mlp_dim=3072, num_layers=12, num_heads=12, pos_embed='conv', classification=False, num_classes=2, dropout_rate=0.0, spatial_dims=3)[source]¶
Vision Transformer (ViT), based on: “Dosovitskiy et al., An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale <https://arxiv.org/abs/2010.11929>”
- Parameters
in_channels (
int
) – dimension of input channels.img_size (
Union
[Sequence
[int
],int
]) – dimension of input image.patch_size (
Union
[Sequence
[int
],int
]) – dimension of patch size.hidden_size (
int
) – dimension of hidden layer.mlp_dim (
int
) – dimension of feedforward layer.num_layers (
int
) – number of transformer blocks.num_heads (
int
) – number of attention heads.pos_embed (
str
) – position embedding layer type.classification (
bool
) – bool argument to determine if classification is used.num_classes (
int
) – number of classes if classification is used.dropout_rate (
float
) – faction of the input units to drop.spatial_dims (
int
) – number of spatial dimensions.
Examples:
# for single channel input with image size of (96,96,96), conv position embedding and segmentation backbone >>> net = ViT(in_channels=1, img_size=(96,96,96), pos_embed='conv') # for 3-channel with image size of (128,128,128), 24 layers and classification backbone >>> net = ViT(in_channels=3, img_size=(128,128,128), pos_embed='conv', classification=True) # for 3-channel with image size of (224,224), 12 layers and classification backbone >>> net = ViT(in_channels=3, img_size=(224,224), pos_embed='conv', classification=True, spatial_dims=2)
- __init__(in_channels, img_size, patch_size, hidden_size=768, mlp_dim=3072, num_layers=12, num_heads=12, pos_embed='conv', classification=False, num_classes=2, dropout_rate=0.0, spatial_dims=3)[source]¶
- Parameters
in_channels (
int
) – dimension of input channels.img_size (
Union
[Sequence
[int
],int
]) – dimension of input image.patch_size (
Union
[Sequence
[int
],int
]) – dimension of patch size.hidden_size (
int
) – dimension of hidden layer.mlp_dim (
int
) – dimension of feedforward layer.num_layers (
int
) – number of transformer blocks.num_heads (
int
) – number of attention heads.pos_embed (
str
) – position embedding layer type.classification (
bool
) – bool argument to determine if classification is used.num_classes (
int
) – number of classes if classification is used.dropout_rate (
float
) – faction of the input units to drop.spatial_dims (
int
) – number of spatial dimensions.
Examples:
# for single channel input with image size of (96,96,96), conv position embedding and segmentation backbone >>> net = ViT(in_channels=1, img_size=(96,96,96), pos_embed='conv') # for 3-channel with image size of (128,128,128), 24 layers and classification backbone >>> net = ViT(in_channels=3, img_size=(128,128,128), pos_embed='conv', classification=True) # for 3-channel with image size of (224,224), 12 layers and classification backbone >>> net = ViT(in_channels=3, img_size=(224,224), pos_embed='conv', classification=True, spatial_dims=2)
- forward(x)[source]¶
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.
ViTAutoEnc¶
- class monai.networks.nets.ViTAutoEnc(in_channels, img_size, patch_size, out_channels=1, deconv_chns=16, hidden_size=768, mlp_dim=3072, num_layers=12, num_heads=12, pos_embed='conv', dropout_rate=0.0, spatial_dims=3)[source]¶
Vision Transformer (ViT), based on: “Dosovitskiy et al., An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale <https://arxiv.org/abs/2010.11929>”
Modified to also give same dimension outputs as the input size of the image
- Parameters
in_channels (
int
) – dimension of input channels or the number of channels for inputimg_size (
Union
[Sequence
[int
],int
]) – dimension of input image.patch_size (
Union
[Sequence
[int
],int
]) – dimension of patch size.hidden_size (
int
) – dimension of hidden layer.out_channels (
int
) – number of output channels.deconv_chns (
int
) – number of channels for the deconvolution layers.mlp_dim (
int
) – dimension of feedforward layer.num_layers (
int
) – number of transformer blocks.num_heads (
int
) – number of attention heads.pos_embed (
str
) – position embedding layer type.dropout_rate (
float
) – faction of the input units to drop.spatial_dims (
int
) – number of spatial dimensions.
Examples:
# for single channel input with image size of (96,96,96), conv position embedding and segmentation backbone # It will provide an output of same size as that of the input >>> net = ViTAutoEnc(in_channels=1, patch_size=(16,16,16), img_size=(96,96,96), pos_embed='conv') # for 3-channel with image size of (128,128,128), output will be same size as of input >>> net = ViTAutoEnc(in_channels=3, patch_size=(16,16,16), img_size=(128,128,128), pos_embed='conv')
- __init__(in_channels, img_size, patch_size, out_channels=1, deconv_chns=16, hidden_size=768, mlp_dim=3072, num_layers=12, num_heads=12, pos_embed='conv', dropout_rate=0.0, spatial_dims=3)[source]¶
- Parameters
in_channels (
int
) – dimension of input channels or the number of channels for inputimg_size (
Union
[Sequence
[int
],int
]) – dimension of input image.patch_size (
Union
[Sequence
[int
],int
]) – dimension of patch size.hidden_size (
int
) – dimension of hidden layer.out_channels (
int
) – number of output channels.deconv_chns (
int
) – number of channels for the deconvolution layers.mlp_dim (
int
) – dimension of feedforward layer.num_layers (
int
) – number of transformer blocks.num_heads (
int
) – number of attention heads.pos_embed (
str
) – position embedding layer type.dropout_rate (
float
) – faction of the input units to drop.spatial_dims (
int
) – number of spatial dimensions.
Examples:
# for single channel input with image size of (96,96,96), conv position embedding and segmentation backbone # It will provide an output of same size as that of the input >>> net = ViTAutoEnc(in_channels=1, patch_size=(16,16,16), img_size=(96,96,96), pos_embed='conv') # for 3-channel with image size of (128,128,128), output will be same size as of input >>> net = ViTAutoEnc(in_channels=3, patch_size=(16,16,16), img_size=(128,128,128), pos_embed='conv')
FullyConnectedNet¶
- class monai.networks.nets.FullyConnectedNet(in_channels, out_channels, hidden_channels, dropout=None, act='PRELU', bias=True, adn_ordering=None)[source]¶
Simple full-connected layer neural network composed of a sequence of linear layers with PReLU activation and dropout. The network accepts input with in_channels channels, has output with out_channels channels, and hidden layer output channels given in hidden_channels. If bias is True then linear units have a bias term.
- Parameters
in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.hidden_channels (
Sequence
[int
]) – number of output channels for each hidden layer.dropout (
Union
[Tuple
,str
,float
,None
]) – dropout ratio. Defaults to no dropout.act (
Union
[Tuple
,str
,None
]) – activation type and arguments. Defaults to PReLU.bias (
bool
) – whether to have a bias term in linear units. Defaults to True.adn_ordering (
Optional
[str
]) – order of operations inmonai.networks.blocks.ADN
.
Examples:
# accepts 4 values and infers 3 values as output, has 3 hidden layers with 10, 20, 10 values as output net = FullyConnectedNet(4, 3, [10, 20, 10], dropout=0.2)
Defines a network accept input with in_channels channels, output of out_channels channels, and hidden layers with channels given in hidden_channels. If bias is True then linear units have a bias term.
- __init__(in_channels, out_channels, hidden_channels, dropout=None, act='PRELU', bias=True, adn_ordering=None)[source]¶
Defines a network accept input with in_channels channels, output of out_channels channels, and hidden layers with channels given in hidden_channels. If bias is True then linear units have a bias term.
VarFullyConnectedNet¶
- class monai.networks.nets.VarFullyConnectedNet(in_channels, out_channels, latent_size, encode_channels, decode_channels, dropout=None, act='PRELU', bias=True, adn_ordering=None)[source]¶
Variational fully-connected network. This is composed of an encode layer, reparameterization layer, and then a decode layer.
- Parameters
in_channels (
int
) – number of input channels.out_channels (
int
) – number of output channels.latent_size (
int
) – number of latent variables to use.encode_channels (
Sequence
[int
]) – number of output channels for each hidden layer of the encode half.decode_channels (
Sequence
[int
]) – number of output channels for each hidden layer of the decode half.dropout (
Union
[Tuple
,str
,float
,None
]) – dropout ratio. Defaults to no dropout.act (
Union
[Tuple
,str
,None
]) – activation type and arguments. Defaults to PReLU.bias (
bool
) – whether to have a bias term in linear units. Defaults to True.adn_ordering (
Optional
[str
]) – order of operations inmonai.networks.blocks.ADN
.
Examples:
# accepts inputs with 4 values, uses a latent space of 2 variables, and produces outputs of 3 values net = VarFullyConnectedNet(4, 3, 2, [5, 10], [10, 5])
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Tuple
[Tensor
,Tensor
,Tensor
,Tensor
]
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)[source]¶
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
monai.networks.blocks.Convolution
ormonai.networks.blocks.ResidualUnit
layers.The layers are constructed using the values in the channels and strides arguments, the number being defined by the length of these (which must match). Input is first passed through a
torch.nn.Linear
layer to convert the input vector to an image tensor with dimensions start_shape. This passes through the convolution layers and is progressively upsampled if the strides values are greater than 1 using transpose convolutions. The size of the final output is defined by the start_shape dimension and the amount of upsampling done through strides. In the default definition the size of the output’s spatial dimensions will be that of start_shape multiplied by the product of strides, thus the example network below upsamples an starting size of (64, 8, 8) to (1, 64, 64) since its strides are (2, 2, 2).- Parameters
latent_shape (
Sequence
[int
]) – tuple of integers stating the dimension of the input latent vector (minus batch dimension)start_shape (
Sequence
[int
]) – tuple of integers stating the dimension of the tensor to pass to convolution subnetworkchannels (
Sequence
[int
]) – tuple of integers stating the output channels of each convolutional layerstrides (
Sequence
[int
]) – tuple of integers stating the stride (upscale factor) of each convolutional layerkernel_size (
Union
[Sequence
[int
],int
]) – integer or tuple of integers stating size of convolutional kernelsnum_res_units (
int
) – integer stating number of convolutions in residual units, 0 means no residual unitsact – name or type defining activation layers
norm – name or type defining normalization layers
dropout (
Optional
[float
]) – optional float value in range [0, 1] stating dropout probability for layers, None for no dropoutbias (
bool
) – boolean stating if convolution layers should have a bias component
Examples:
# 3 layers, latent input vector of shape (42, 24), output volume of shape (1, 64, 64) net = Generator((42, 24), (64, 8, 8), (32, 16, 1), (2, 2, 2))
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Tensor
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)[source]¶
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.
The network is constructed as a sequence of layers, either
monai.networks.blocks.Convolution
ormonai.networks.blocks.ResidualUnit
, with a final fully-connected layer resizing the output from the blocks to the final size. Each block is defined with a stride value typically used to downsample the input using strided convolutions. In this way each block progressively condenses information from the input into a deep representation the final fully-connected layer relates to a final result.- Parameters
in_shape (
Sequence
[int
]) – tuple of integers stating the dimension of the input tensor (minus batch dimension)out_shape (
Sequence
[int
]) – tuple of integers stating the dimension of the final output tensor (minus batch dimension)channels (
Sequence
[int
]) – tuple of integers stating the output channels of each convolutional layerstrides (
Sequence
[int
]) – tuple of integers stating the stride (downscale factor) of each convolutional layerkernel_size (
Union
[Sequence
[int
],int
]) – integer or tuple of integers stating size of convolutional kernelsnum_res_units (
int
) – integer stating number of convolutions in residual units, 0 means no residual unitsact – name or type defining activation layers
norm – name or type defining normalization layers
dropout (
Optional
[float
]) – optional float value in range [0, 1] stating dropout probability for layers, None for no dropoutbias (
bool
) – boolean stating if convolution layers should have a bias component
Examples:
# infers a 2-value result (eg. a 2D cartesian coordinate) from a 64x64 image net = Regressor((1, 64, 64), (2,), (2, 4, 8), (2, 2, 2))
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Tensor
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)[source]¶
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.
- Parameters
in_shape (
Sequence
[int
]) – tuple of integers stating the dimension of the input tensor (minus batch dimension)classes (
int
) – integer stating the dimension of the final output tensorchannels (
Sequence
[int
]) – tuple of integers stating the output channels of each convolutional layerstrides (
Sequence
[int
]) – tuple of integers stating the stride (downscale factor) of each convolutional layerkernel_size (
Union
[Sequence
[int
],int
]) – integer or tuple of integers stating size of convolutional kernelsnum_res_units (
int
) – integer stating number of convolutions in residual units, 0 means no residual unitsact – name or type defining activation layers
norm – name or type defining normalization layers
dropout (
Optional
[float
]) – optional float value in range [0, 1] stating dropout probability for layers, None for no dropoutbias (
bool
) – boolean stating if convolution layers should have a bias componentlast_act (
Optional
[str
]) – name defining the last activation layer
Initializes internal Module state, shared by both nn.Module and ScriptModule.
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')[source]¶
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.
- Parameters
in_shape (
Sequence
[int
]) – tuple of integers stating the dimension of the input tensor (minus batch dimension)channels (
Sequence
[int
]) – tuple of integers stating the output channels of each convolutional layerstrides (
Sequence
[int
]) – tuple of integers stating the stride (downscale factor) of each convolutional layerkernel_size (
Union
[Sequence
[int
],int
]) – integer or tuple of integers stating size of convolutional kernelsnum_res_units (
int
) – integer stating number of convolutions in residual units, 0 means no residual unitsact – name or type defining activation layers
norm – name or type defining normalization layers
dropout (
Optional
[float
]) – optional float value in range [0, 1] stating dropout probability for layers, None for no dropoutbias (
bool
) – boolean stating if convolution layers should have a bias componentlast_act – name defining the last activation layer
Initializes internal Module state, shared by both nn.Module and ScriptModule.
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)[source]¶
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 Wasserstein GANs.
- Parameters
in_shape (
Sequence
[int
]) – tuple of integers stating the dimension of the input tensor (minus batch dimension)channels (
Sequence
[int
]) – tuple of integers stating the output channels of each convolutional layerstrides (
Sequence
[int
]) – tuple of integers stating the stride (downscale factor) of each convolutional layerkernel_size (
Union
[Sequence
[int
],int
]) – integer or tuple of integers stating size of convolutional kernelsnum_res_units (
int
) – integer stating number of convolutions in residual units, 0 means no residual unitsact – name or type defining activation layers
norm – name or type defining normalization layers
dropout (
Optional
[float
]) – optional float value in range [0, 1] stating dropout probability for layers, None for no dropoutbias (
bool
) – boolean stating if convolution layers should have a bias component
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.- Return type
Tensor
Transchex¶
- class monai.networks.nets.Transchex(in_channels, img_size, patch_size, num_classes, num_language_layers, num_vision_layers, num_mixed_layers, hidden_size=768, drop_out=0.0, attention_probs_dropout_prob=0.1, gradient_checkpointing=False, hidden_act='gelu', hidden_dropout_prob=0.1, initializer_range=0.02, intermediate_size=3072, layer_norm_eps=1e-12, max_position_embeddings=512, model_type='bert', num_attention_heads=12, num_hidden_layers=12, pad_token_id=0, position_embedding_type='absolute', transformers_version='4.10.2', type_vocab_size=2, use_cache=True, vocab_size=30522, chunk_size_feed_forward=0, is_decoder=False, add_cross_attention=False)[source]¶
TransChex based on: “Hatamizadeh et al.,TransCheX: Self-Supervised Pretraining of Vision-Language Transformers for Chest X-ray Analysis”
- Parameters
in_channels (
int
) – dimension of input channels.img_size (
Union
[Sequence
[int
],int
]) – dimension of input image.patch_size (
Union
[int
,Tuple
[int
,int
]]) – dimension of patch size.num_classes (
int
) – number of classes if classification is used.num_language_layers (
int
) – number of language transformer layers.num_vision_layers (
int
) – number of vision transformer layers.num_mixed_layers (
int
) – number of mixed transformer layers.drop_out (
float
) – faction of the input units to drop.
The other parameters are part of the bert_config to MultiModal.from_pretrained.
Examples:
# for 3-channel with image size of (224,224), patch size of (32,32), 3 classes, 2 language layers, # 2 vision layers, 2 mixed modality layers and dropout of 0.2 in the classification head net = Transchex(in_channels=3, img_size=(224, 224), num_classes=3, num_language_layers=2, num_vision_layers=2, num_mixed_layers=2, drop_out=0.2)
- __init__(in_channels, img_size, patch_size, num_classes, num_language_layers, num_vision_layers, num_mixed_layers, hidden_size=768, drop_out=0.0, attention_probs_dropout_prob=0.1, gradient_checkpointing=False, hidden_act='gelu', hidden_dropout_prob=0.1, initializer_range=0.02, intermediate_size=3072, layer_norm_eps=1e-12, max_position_embeddings=512, model_type='bert', num_attention_heads=12, num_hidden_layers=12, pad_token_id=0, position_embedding_type='absolute', transformers_version='4.10.2', type_vocab_size=2, use_cache=True, vocab_size=30522, chunk_size_feed_forward=0, is_decoder=False, add_cross_attention=False)[source]¶
- Parameters
in_channels (
int
) – dimension of input channels.img_size (
Union
[Sequence
[int
],int
]) – dimension of input image.patch_size (
Union
[int
,Tuple
[int
,int
]]) – dimension of patch size.num_classes (
int
) – number of classes if classification is used.num_language_layers (
int
) – number of language transformer layers.num_vision_layers (
int
) – number of vision transformer layers.num_mixed_layers (
int
) – number of mixed transformer layers.drop_out (
float
) – faction of the input units to drop.
The other parameters are part of the bert_config to MultiModal.from_pretrained.
Examples:
# for 3-channel with image size of (224,224), patch size of (32,32), 3 classes, 2 language layers, # 2 vision layers, 2 mixed modality layers and dropout of 0.2 in the classification head net = Transchex(in_channels=3, img_size=(224, 224), num_classes=3, num_language_layers=2, num_vision_layers=2, num_mixed_layers=2, drop_out=0.2)
- forward(input_ids, token_type_ids=None, vision_feats=None)[source]¶
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.
NetAdapter¶
- class monai.networks.nets.NetAdapter(model, num_classes=1, dim=2, in_channels=None, use_conv=False, pool=('avg', {'kernel_size': 7, 'stride': 1}), bias=True, n_classes=None)[source]¶
Wrapper to replace the last layer of model by convolutional layer or FC layer. This module expects the output of model layers[0: -2] is a feature map with shape [B, C, spatial dims], then replace the model’s last two layers with an optional pooling and a conv or linear layer.
- Parameters
model (
Module
) – a PyTorch model, which can be both 2D and 3D models. typically, it can be a pretrained model in Torchvision, like:resnet18
,resnet34
,resnet50
,resnet101
,resnet152
, etc. more details: https://pytorch.org/vision/stable/models.html.num_classes (
int
) – number of classes for the last classification layer. Default to 1.dim (
int
) – number of supported spatial dimensions in the specified model, depends on the model implementation. default to 2 as most Torchvision models are for 2D image processing.in_channels (
Optional
[int
]) – number of the input channels of last layer. if None, get it from in_features of last layer.use_conv (
bool
) – whether use convolutional layer to replace the last layer, default to False.pool (
Optional
[Tuple
[str
,Dict
[str
,Any
]]]) – parameters for the pooling layer, it should be a tuple, the first item is name of the pooling layer, the second item is dictionary of the initialization args. if None, will not replace the layers[-2]. default to (“avg”, {“kernel_size”: 7, “stride”: 1}).bias (
bool
) – the bias value when replacing the last layer. if False, the layer will not learn an additive bias, default to True.
Deprecated since version 0.6.0:
n_classes
is deprecated, usenum_classes
instead.Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x)[source]¶
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.
TorchVisionFCModel¶
- class monai.networks.nets.TorchVisionFCModel(model_name='resnet18', num_classes=1, dim=2, in_channels=None, use_conv=False, pool=('avg', {'kernel_size': 7, 'stride': 1}), bias=True, pretrained=False, n_classes=None)[source]¶
Customize the fully connected layer of TorchVision model or replace it by convolutional layer.
- Parameters
model_name (
str
) – name of any torchvision model with fully connected layer at the end.resnet18
(default),resnet34
,resnet50
,resnet101
,resnet152
,resnext50_32x4d
,resnext101_32x8d
,wide_resnet50_2
,wide_resnet101_2
. model details: https://pytorch.org/vision/stable/models.html.num_classes (
int
) – number of classes for the last classification layer. Default to 1.dim (
int
) – number of supported spatial dimensions in the specified model, depends on the model implementation. default to 2 as most Torchvision models are for 2D image processing.in_channels (
Optional
[int
]) – number of the input channels of last layer. if None, get it from in_features of last layer.use_conv (
bool
) – whether use convolutional layer to replace the last layer, default to False.pool (
Optional
[Tuple
[str
,Dict
[str
,Any
]]]) – parameters for the pooling layer, it should be a tuple, the first item is name of the pooling layer, the second item is dictionary of the initialization args. if None, will not replace the layers[-2]. default to (“avg”, {“kernel_size”: 7, “stride”: 1}).bias (
bool
) – the bias value when replacing the last layer. if False, the layer will not learn an additive bias, default to True.pretrained (
bool
) – whether to use the imagenet pretrained weights. Default to False.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
TorchVisionFullyConvModel¶
- class monai.networks.nets.TorchVisionFullyConvModel(model_name='resnet18', num_classes=1, pool_size=(7, 7), pool_stride=1, pretrained=False, n_classes=None)[source]¶
Customize TorchVision models to replace fully connected layer by convolutional layer.
- Parameters
model_name (
str
) – name of any torchvision with adaptive avg pooling and fully connected layer at the end.resnet18
(default),resnet34
,resnet50
,resnet101
,resnet152
,resnext50_32x4d
,resnext101_32x8d
,wide_resnet50_2
,wide_resnet101_2
.num_classes (
int
) – number of classes for the last classification layer. Default to 1.pool_size (
Union
[int
,Tuple
[int
,int
]]) – the kernel size for AvgPool2d to replace AdaptiveAvgPool2d. Default to (7, 7).pool_stride (
Union
[int
,Tuple
[int
,int
]]) – the stride for AvgPool2d to replace AdaptiveAvgPool2d. Default to 1.pretrained (
bool
) – whether to use the imagenet pretrained weights. Default to False.
Deprecated since version 0.6.0: Use
monai.networks.nets.TorchVisionFCModel
instead.Initializes internal Module state, shared by both nn.Module and ScriptModule.
MILModel¶
- class monai.networks.nets.MILModel(num_classes, mil_mode='att', pretrained=True, backbone=None, backbone_num_features=None, trans_blocks=4, trans_dropout=0.0)[source]¶
Multiple Instance Learning (MIL) model, with a backbone classification model. Currently, it only works for 2D images, a typical use case is for classification of the digital pathology whole slide images. The expected shape of input data is [B, N, C, H, W], where B is the batch_size of PyTorch Dataloader and N is the number of instances extracted from every original image in the batch. A tutorial example is available at: https://github.com/Project-MONAI/tutorials/tree/master/pathology/multiple_instance_learning.
- Parameters
num_classes (
int
) – number of output classes.mil_mode (
str
) –MIL algorithm, available values (Defaults to
"att"
):"mean"
- average features from all instances, equivalent to pure CNN (non MIL)."max"
- retain only the instance with the max probability for loss calculation."att"
- attention based MIL https://arxiv.org/abs/1802.04712."att_trans"
- transformer MIL https://arxiv.org/abs/2111.01556."att_trans_pyramid"
- transformer pyramid MIL https://arxiv.org/abs/2111.01556.
pretrained (
bool
) – init backbone with pretrained weights, defaults toTrue
.backbone (
Union
[Module
,str
,None
]) – Backbone classifier CNN (eitherNone
, ann.Module
that returns features, or a string name of a torchvision model). Defaults toNone
, in which case ResNet50 is used.backbone_num_features (
Optional
[int
]) – Number of output features of the backbone CNN Defaults toNone
(necessary only when using a custom backbone)trans_blocks (
int
) – number of the blocks in TransformEncoder layer.trans_dropout (
float
) – dropout rate in TransformEncoder layer.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
- forward(x, no_head=False)[source]¶
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.- Return type
Tensor
DiNTS¶
- class monai.networks.nets.DiNTS(dints_space, in_channels, num_classes, act_name='RELU', norm_name='INSTANCE', spatial_dims=3, use_downsample=True, node_a=None)[source]¶
Reimplementation of DiNTS based on “DiNTS: Differentiable Neural Network Topology Search for 3D Medical Image Segmentation <https://arxiv.org/abs/2103.15954>”.
The model contains a pre-defined multi-resolution stem block (defined in this class) and a DiNTS space (defined in
monai.networks.nets.TopologyInstance
andmonai.networks.nets.TopologySearch
).The stem block is for: 1) input downsample and 2) output upsample to original size. The model downsamples the input image by 2 (if
use_downsample=True
). The downsampled image is downsampled by [1, 2, 4, 8] times (num_depths=4
) and used as input to the DiNTS search space (TopologySearch
) or the DiNTS instance (TopologyInstance
).TopologyInstance
is the final searched model. The initialization requires the searched architecture codes.TopologySearch
is a multi-path topology and cell operation search space. The architecture codes will be initialized as one.TopologyConstruction
is the parent class which constructs the instance and search space.
To meet the requirements of the structure, the input size for each spatial dimension should be: divisible by 2 ** (num_depths + 1).
- Parameters
dints_space – DiNTS search space. The value should be instance of TopologyInstance or TopologySearch.
in_channels (
int
) – number of input image channels.num_classes (
int
) – number of output segmentation classes.act_name (
Union
[Tuple
,str
]) – activation name, default to ‘RELU’.norm_name (
Union
[Tuple
,str
]) – normalization used in convolution blocks. Default to InstanceNorm.spatial_dims (
int
) – spatial 2D or 3D inputs.use_downsample (
bool
) – use downsample in the stem. IfFalse
, the search space will be in resolution [1, 1/2, 1/4, 1/8], ifTrue
, the search space will be in resolution [1/2, 1/4, 1/8, 1/16].node_a – node activation numpy matrix. Its shape is (num_depths, num_blocks + 1). +1 for multi-resolution inputs. In model searching stage,
node_a
can be None. In deployment stage,node_a
cannot be None.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
TopologyConstruction for DiNTS¶
- class monai.networks.nets.TopologyConstruction(arch_code=None, channel_mul=1.0, cell=<class 'monai.networks.nets.dints.Cell'>, num_blocks=6, num_depths=3, spatial_dims=3, use_downsample=True, device='cpu')[source]¶
The base class for TopologyInstance and TopologySearch.
- Parameters
arch_code (
Optional
[list
]) –[arch_code_a, arch_code_c], numpy arrays. The architecture codes defining the model. For example, for a
num_depths=4, num_blocks=12
search space:arch_code_a is a 12x10 (10 paths) binary matrix representing if a path is activated.
arch_code_c is a 12x10x5 (5 operations) binary matrix representing if a cell operation is used.
arch_code in
__init__()
is used for creating the network and remove unused network blocks. If None,
all paths and cells operations will be used, and must be in the searching stage (is_search=True).
channel_mul (
float
) – adjust intermediate channel number, default is 1.cell – operation of each node.
num_blocks (
int
) – number of blocks (depth in the horizontal direction) of the DiNTS search space.num_depths (
int
) – number of image resolutions of the DiNTS search space: 1, 1/2, 1/4 … in each dimension.use_downsample (
bool
) – use downsample in the stem. If False, the search space will be in resolution [1, 1/2, 1/4, 1/8], if True, the search space will be in resolution [1/2, 1/4, 1/8, 1/16].device (
str
) – ‘cpu’, ‘cuda’, or device ID.
- Predefined variables:
filter_nums: default to 32. Double the number of channels after downsample. topology related variables:
arch_code2in: path activation to its incoming node index (resolution). For depth = 4, arch_code2in = [0, 1, 0, 1, 2, 1, 2, 3, 2, 3]. The first path outputs from node 0 (top resolution), the second path outputs from node 1 (second resolution in the search space), the third path outputs from node 0, etc.
arch_code2ops: path activation to operations of upsample 1, keep 0, downsample -1. For depth = 4, arch_code2ops = [0, 1, -1, 0, 1, -1, 0, 1, -1, 0]. The first path does not change resolution, the second path perform upsample, the third perform downsample, etc.
arch_code2out: path activation to its output node index. For depth = 4, arch_code2out = [0, 0, 1, 1, 1, 2, 2, 2, 3, 3], the first and second paths connects to node 0 (top resolution), the 3,4,5 paths connects to node 1, etc.
Initializes internal Module state, shared by both nn.Module and ScriptModule.
TopologyInstance for DiNTS¶
- class monai.networks.nets.TopologyInstance(arch_code=None, channel_mul=1.0, cell=<class 'monai.networks.nets.dints.Cell'>, num_blocks=6, num_depths=3, spatial_dims=3, use_downsample=True, device='cpu')[source]¶
Instance of the final searched architecture. Only used in re-training/inference stage.
Initialize DiNTS topology search space of neural architectures.
TopologySearch for DiNTS¶
- class monai.networks.nets.TopologySearch(channel_mul=1.0, cell=<class 'monai.networks.nets.dints.Cell'>, arch_code=None, num_blocks=6, num_depths=3, spatial_dims=3, use_downsample=True, device='cpu')[source]¶
DiNTS topology search space of neural architectures.
Examples:
from monai.networks.nets.dints import TopologySearch topology_search_space = TopologySearch( channel_mul=0.5, num_blocks=8, num_depths=4, use_downsample=True, spatial_dims=3) topology_search_space.get_ram_cost_usage(in_size=(2, 16, 80, 80, 80), full=True) multi_res_images = [ torch.randn(2, 16, 80, 80, 80), torch.randn(2, 32, 40, 40, 40), torch.randn(2, 64, 20, 20, 20), torch.randn(2, 128, 10, 10, 10)] prediction = topology_search_space(image) for x in prediction: print(x.shape) # torch.Size([2, 16, 80, 80, 80]) # torch.Size([2, 32, 40, 40, 40]) # torch.Size([2, 64, 20, 20, 20]) # torch.Size([2, 128, 10, 10, 10])
Class method overview:
get_prob_a()
: convert learnable architecture weights to path activation probabilities.get_ram_cost_usage()
: get estimated ram cost.get_topology_entropy()
: get topology entropy loss in searching stage.decode()
: get final binarized architecture code.gen_mtx()
: generate variables needed for topology search.
- Predefined variables:
tidx: index used to convert path activation matrix T = (depth,depth) in transfer_mtx to path activation arch_code (1,3*depth-2), for depth = 4, tidx = [0, 1, 4, 5, 6, 9, 10, 11, 14, 15], A tidx (10 binary values) represents the path activation.
transfer_mtx: feasible path activation matrix (denoted as T) given a node activation pattern. It is used to convert path activation pattern (1, paths) to node activation (1, nodes)
node_act_list: all node activation [2^num_depths-1, depth]. For depth = 4, there are 15 node activation patterns, each of length 4. For example, [1,1,0,0] means nodes 0, 1 are activated (with input paths).
all_connect: All possible path activations. For depth = 4, all_connection has 1024 vectors of length 10 (10 paths). The return value will exclude path activation of all 0.
Initialize DiNTS topology search space of neural architectures.
- __init__(channel_mul=1.0, cell=<class 'monai.networks.nets.dints.Cell'>, arch_code=None, num_blocks=6, num_depths=3, spatial_dims=3, use_downsample=True, device='cpu')[source]¶
Initialize DiNTS topology search space of neural architectures.
- decode()[source]¶
Decode network log_alpha_a/log_alpha_c using dijkstra shortest path algorithm.
[node_a, arch_code_a, arch_code_c, arch_code_a_max] is decoded when using
self.decode()
.For example, for a
num_depths=4
,num_blocks=12
search space:node_a
is a 4x13 binary matrix representing if a feature node is activated (13 because of multi-resolution inputs).arch_code_a
is a 12x10 (10 paths) binary matrix representing if a path is activated.arch_code_c
is a 12x10x5 (5 operations) binary matrix representing if a cell operation is used.
- Returns
arch_code with maximum probability
- forward(x)[source]¶
Prediction based on dynamic arch_code.
- Parameters
x – a list of num_depths input tensors as a multi-resolution input. tensor is of shape BCHW[D] where C must match self.filter_nums.
- gen_mtx(depth)[source]¶
Generate elements needed in decoding and topology.
- transfer_mtx: feasible path activation matrix (denoted as T) given a node activation pattern.
It is used to convert path activation pattern (1, paths) to node activation (1, nodes)
- node_act_list: all node activation [2^num_depths-1, depth]. For depth = 4, there are 15 node activation
patterns, each of length 4. For example, [1,1,0,0] means nodes 0, 1 are activated (with input paths).
all_connect: All possible path activations. For depth = 4, all_connection has 1024 vectors of length 10 (10 paths). The return value will exclude path activation of all 0.
- get_prob_a(child=False)[source]¶
Get final path and child model probabilities from architecture weights log_alpha_a. This is used in forward pass, getting training loss, and final decoding.
- Parameters
child (
bool
) – return child probability (used in decoding)- Returns
- the path activation probability of size:
[number of blocks, number of paths in each block]. For 12 blocks, 4 depths search space, the size is [12,10]
- probs_a: The probability of all child models (size 1023x10). Each child model is a path activation pattern
(1D vector of length 10 for 10 paths). In total 1023 child models (2^10 -1)
- Return type
arch_code_prob_a
Utilities¶
Utilities and types for defining networks, these depend on PyTorch.
- monai.networks.utils.convert_to_torchscript(model, filename_or_obj=None, extra_files=None, verify=False, inputs=None, device=None, rtol=0.0001, atol=0.0, **kwargs)[source]¶
Utility to convert a model into TorchScript model and save to file, with optional input / output data verification.
- Parameters
model (
Module
) – source PyTorch model to save.filename_or_obj (
Optional
[Any
]) – if not None, specify a file-like object (has to implement write and flush) or a string containing a file path name to save the TorchScript model.extra_files (
Optional
[Dict
]) – map from filename to contents which will be stored as part of the save model file. works for PyTorch 1.7 or later. for more details: https://pytorch.org/docs/stable/generated/torch.jit.save.html.verify (
bool
) – whether to verify the input and output of TorchScript model. if filename_or_obj is not None, load the saved TorchScript model and verify.inputs (
Optional
[Sequence
[Any
]]) – input test data to verify model, should be a sequence of data, every item maps to a argument of model() function.device (
Optional
[device
]) – target device to verify the model, if None, use CUDA if available.rtol (
float
) – the relative tolerance when comparing the outputs of PyTorch model and TorchScript model.atol (
float
) – the absolute tolerance when comparing the outputs of PyTorch model and TorchScript model.
- monai.networks.utils.copy_model_state(dst, src, dst_prefix='', mapping=None, exclude_vars=None, inplace=True)[source]¶
Compute a module state_dict, of which the keys are the same as dst. The values of dst are overwritten by the ones from src whenever their keys match. The method provides additional dst_prefix for the dst key when matching them. mapping can be a {“src_key”: “dst_key”} dict, indicating dst[dst_prefix + dst_key] = src[src_key]. This function is mainly to return a model state dict for loading the src model state into the dst model, src and dst can have different dict keys, but their corresponding values normally have the same shape.
- Parameters
dst (
Union
[Module
,Mapping
]) – a pytorch module or state dict to be updated.src (
Union
[Module
,Mapping
]) – a pytorch module or state dist used to get the values used for the update.dst_prefix – dst key prefix, so that dst[dst_prefix + src_key] will be assigned to the value of src[src_key].
mapping – a {“src_key”: “dst_key”} dict, indicating that dst[dst_prefix + dst_key] to be assigned to the value of src[src_key].
exclude_vars – a regular expression to match the dst variable names, so that their values are not overwritten by src.
inplace – whether to set the dst module with the updated state_dict via load_state_dict. This option is only available when dst is a torch.nn.Module.
Examples
from monai.networks.nets import BasicUNet from monai.networks.utils import copy_model_state model_a = BasicUNet(in_channels=1, out_channels=4) model_b = BasicUNet(in_channels=1, out_channels=2) model_a_b, changed, unchanged = copy_model_state( model_a, model_b, exclude_vars="conv_0.conv_0", inplace=False) # dst model updated: 76 of 82 variables. model_a.load_state_dict(model_a_b) # <All keys matched successfully>
Returns: an OrderedDict of the updated dst state, the changed, and unchanged keys.
- monai.networks.utils.eval_mode(*nets)[source]¶
Set network(s) to eval mode and then return to original state at the end.
- Parameters
nets (
Module
) – Input network(s)
Examples
t=torch.rand(1,1,16,16) p=torch.nn.Conv2d(1,1,3) print(p.training) # True with eval_mode(p): print(p.training) # False print(p(t).sum().backward()) # will correctly raise an exception as gradients are calculated
- monai.networks.utils.get_state_dict(obj)[source]¶
Get the state dict of input object if has state_dict, otherwise, return object directly. For data parallel model, automatically convert it to regular model first.
- Parameters
obj (
Union
[Module
,Mapping
]) – input object to check and get the state_dict.
- monai.networks.utils.icnr_init(conv, upsample_factor, init=<function kaiming_normal_>)[source]¶
ICNR initialization for 2D/3D kernels adapted from Aitken et al.,2017 , “Checkerboard artifact free sub-pixel convolution”.
- monai.networks.utils.normal_init(m, std=0.02, normal_func=<function normal_>)[source]¶
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.
- Return type
None
- monai.networks.utils.normalize_transform(shape, device=None, dtype=None, align_corners=False)[source]¶
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
[int
]) – input spatial shapedevice (
Optional
[device
]) – device on which the returned affine will be allocated.dtype (
Optional
[dtype
]) – data type of the returned affinealign_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
- Return type
Tensor
- monai.networks.utils.one_hot(labels, num_classes, dtype=torch.float32, dim=1)[source]¶
For every value v in labels, the value in the output will be either 1 or 0. Each vector along the dim-th dimension has the “one-hot” format, i.e., it has a total length of num_classes, with a one and num_class-1 zeros. Note that this will include the background label, thus a binary mask should be treated as having two classes.
- Parameters
labels (
Tensor
) – input tensor of integers to be converted into the ‘one-hot’ format. Internally labels will be converted into integers labels.long().num_classes (
int
) – number of output channels, the corresponding length of labels[dim] will be converted to num_classes from 1.dtype (
dtype
) – the data type of the output one_hot label.dim (
int
) – the dimension to be converted to num_classes channels from 1 channel, should be non-negative number.
Example:
For a tensor labels of dimensions [B]1[spatial_dims], return a tensor of dimensions [B]N[spatial_dims] when num_classes=N number of classes and dim=1.
from monai.networks.utils import one_hot import torch a = torch.randint(0, 2, size=(1, 2, 2, 2)) out = one_hot(a, num_classes=2, dim=0) print(out.shape) # torch.Size([2, 2, 2, 2]) a = torch.randint(0, 2, size=(2, 1, 2, 2, 2)) out = one_hot(a, num_classes=2, dim=1) print(out.shape) # torch.Size([2, 2, 2, 2, 2])
- Return type
Tensor
- monai.networks.utils.pixelshuffle(x, spatial_dims, scale_factor, dimensions=None)[source]¶
Apply pixel shuffle to the tensor x with spatial dimensions spatial_dims and scaling factor scale_factor.
See: Shi et al., 2016, “Real-Time Single Image and Video Super-Resolution Using a nEfficient Sub-Pixel Convolutional Neural Network.”
See: Aitken et al., 2017, “Checkerboard artifact free sub-pixel convolution”.
- Parameters
x (
Tensor
) – Input tensorspatial_dims (
int
) – number of spatial dimensions, typically 2 or 3 for 2D or 3Dscale_factor (
int
) – factor to rescale the spatial dimensions by, must be >=1
Deprecated since version 0.6.0:
dimensions
is deprecated, usespatial_dims
instead.- Return type
Tensor
- Returns
Reshuffled version of x.
- Raises
ValueError – When input channels of x are not divisible by (scale_factor ** spatial_dims)
- monai.networks.utils.predict_segmentation(logits, mutually_exclusive=False, threshold=0.0)[source]¶
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.
- Return type
Any