Multi-dimensional GRU

MDGRU

class mdgru.model_pytorch.mdrnn.mdgru.MDRNN(dropout, spatial_dimensions, kw)[source]

Bases: torch.nn.modules.module.Module

MDRNN class originally designed to handle the sum of cGRU computations resulting in one MDGRU.

_defaults contains initial values for most class attributes. :param dropout: Dropoutrate to be applied (provided as keep rate) :param kw: dict containing the following options.

  • use_dropconnect_x [default: True] Should Dropconnect be applied to the input?
  • use_dropconnect_h [default: True] Should DropConnect be applied to the state?
  • return_cgru_results [default: False] Instead of summing, individual cgru channel results are concatenated.
  • filter_size_x [default: [7]] Convolution kernel size for input.
  • filter_size_h [default: [7]] Convolution kernel size for state.
  • crnn_activation [default: class ‘torch.nn.modules.activation.Tanh’] Activation function to be used for the CRNN.
  • legacy_cgru_addition [default: False] results in worse weight initialization, only use if you know what you are doing!
  • crnn_class [default: class ‘mdgru.model_pytorch.crnn.cgru.CGRUCell’] CRNN class to be used in the MDRNN
  • strides [default: None]
  • name [default: mdgru]
  • num_hidden [default: 100]
  • num_input [default: 6]
Parameters:spatial_dimensions – which dimensions should be processed with a cRNN (by default all of them)
_all_buffers(memo=None)
_apply(fn)
_defaults = {'crnn_activation': {'value': <class 'torch.nn.modules.activation.Tanh'>, 'help': 'Activation function to be used for the CRNN.'}, 'crnn_class': {'value': <class 'mdgru.model_pytorch.crnn.cgru.CGRUCell'>, 'help': 'CRNN class to be used in the MDRNN'}, 'filter_size_h': {'value': [7], 'help': 'Convolution kernel size for state.'}, 'filter_size_x': {'value': [7], 'help': 'Convolution kernel size for input.'}, 'legacy_cgru_addition': {'value': False, 'help': 'results in worse weight initialization, only use if you know what you are doing!'}, 'name': 'mdgru', 'num_hidden': 100, 'num_input': 6, 'return_cgru_results': {'value': False, 'help': 'Instead of summing, individual cgru channel results are concatenated.'}, 'strides': None, 'use_dropconnect_h': {'value': True, 'help': 'Should DropConnect be applied to the state?', 'invert_meaning': 'dont_'}, 'use_dropconnect_x': {'value': True, 'help': 'Should Dropconnect be applied to the input?', 'invert_meaning': 'dont_'}}
_get_name()
_load_from_state_dict(state_dict, prefix, strict, missing_keys, unexpected_keys, error_msgs)

Copies parameters and buffers from state_dict into only this module, but not its descendants. This is called on every submodule in load_state_dict(). Metadata saved for this module in input state_dict is at state_dict._metadata[prefix]. Subclasses can achieve class-specific backward compatible loading using the version number at state_dict._metadata[prefix]["version"].

Note

state_dict is not the same object as the input state_dict to load_state_dict(). So it can be modified.

Parameters:
  • state_dict (dict) – a dict containing parameters and persistent buffers.
  • prefix (str) – the prefix for parameters and buffers used in this module
  • strict (bool) – whether to strictly enforce that the keys in state_dict with prefix match the names of parameters and buffers in this module
  • missing_keys (list of str) – if strict=False, add missing keys to this list
  • unexpected_keys (list of str) – if strict=False, add unexpected keys to this list
  • error_msgs (list of str) – error messages should be added to this list, and will be reported together in load_state_dict()
_slow_forward(*input, **kwargs)
_tracing_name(tracing_state)
_version = 1
add_module(name, module)

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Parameters:
  • name (string) – name of the child module. The child module can be accessed from this module using the given name
  • parameter (Module) – child module to be added to the module.
apply(fn)

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also torch-nn-init).

Parameters:fn (Module -> None) – function to be applied to each submodule
Returns:Module – self

Example:

>>> def init_weights(m):
        print(m)
        if type(m) == nn.Linear:
            m.weight.data.fill_(1.0)
            print(m.weight)

>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[ 1.,  1.],
        [ 1.,  1.]])
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[ 1.,  1.],
        [ 1.,  1.]])
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
children()

Returns an iterator over immediate children modules.

Yields:Module – a child module
cpu()

Moves all model parameters and buffers to the CPU.

Returns:Module – self
cuda(device=None)

Moves all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized.

Parameters:device (int, optional) – if specified, all parameters will be copied to that device
Returns:Module – self
double()

Casts all floating point parameters and buffers to double datatype.

Returns:Module – self
dump_patches = False
eval()

Sets the module in evaluation mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

extra_repr()

Set the extra representation of the module

To print customized extra information, you should reimplement this method in your own modules. Both single-line and multi-line strings are acceptable.

float()

Casts all floating point parameters and buffers to float datatype.

Returns:Module – self
forward(input)[source]
half()

Casts all floating point parameters and buffers to half datatype.

Returns:Module – self
load_state_dict(state_dict, strict=True)

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Parameters:
  • state_dict (dict) – a dict containing parameters and persistent buffers.
  • strict (bool, optional) – whether to strictly enforce that the keys in state_dict match the keys returned by this module’s state_dict() function. Default: True
modules()

Returns an iterator over all modules in the network.

Yields:Module – a module in the network

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.modules()):
        print(idx, '->', m)

0 -> Sequential (
  (0): Linear (2 -> 2)
  (1): Linear (2 -> 2)
)
1 -> Linear (2 -> 2)
named_children()

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields:(string, Module) – Tuple containing a name and child module

Example:

>>> for name, module in model.named_children():
>>>     if name in ['conv4', 'conv5']:
>>>         print(module)
named_modules(memo=None, prefix='')

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Yields:(string, Module) – Tuple of name and module

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.named_modules()):
        print(idx, '->', m)

0 -> ('', Sequential (
  (0): Linear (2 -> 2)
  (1): Linear (2 -> 2)
))
1 -> ('0', Linear (2 -> 2))
named_parameters(memo=None, prefix='')

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself

Yields:(string, Parameter) – Tuple containing the name and parameter

Example:

>>> for name, param in self.named_parameters():
>>>    if name in ['bias']:
>>>        print(param.size())
parameters()

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Yields:Parameter – module parameter

Example:

>>> for param in model.parameters():
>>>     print(type(param.data), param.size())
<class 'torch.FloatTensor'> (20L,)
<class 'torch.FloatTensor'> (20L, 1L, 5L, 5L)
register_backward_hook(hook)

Registers a backward hook on the module.

The hook will be called every time the gradients with respect to module inputs are computed. The hook should have the following signature:

hook(module, grad_input, grad_output) -> Tensor or None

The grad_input and grad_output may be tuples if the module has multiple inputs or outputs. The hook should not modify its arguments, but it can optionally return a new gradient with respect to input that will be used in place of grad_input in subsequent computations.

Returns:torch.utils.hooks.RemovableHandle – a handle that can be used to remove the added hook by calling handle.remove()
register_buffer(name, tensor)

Adds a persistent buffer to the module.

This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the persistent state.

Buffers can be accessed as attributes using given names.

Parameters:
  • name (string) – name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor) – buffer to be registered.

Example:

>>> self.register_buffer('running_mean', torch.zeros(num_features))
register_forward_hook(hook)

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output. It should have the following signature:

hook(module, input, output) -> None

The hook should not modify the input or output.

Returns:torch.utils.hooks.RemovableHandle – a handle that can be used to remove the added hook by calling handle.remove()
register_forward_pre_hook(hook)

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked. It should have the following signature:

hook(module, input) -> None

The hook should not modify the input.

Returns:torch.utils.hooks.RemovableHandle – a handle that can be used to remove the added hook by calling handle.remove()
register_parameter(name, param)

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Parameters:
  • name (string) – name of the parameter. The parameter can be accessed from this module using the given name
  • parameter (Parameter) – parameter to be added to the module.
share_memory()
state_dict(destination=None, prefix='', keep_vars=False)

Returns a dictionary containing a whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names.

Returns:dict – a dictionary containing a whole state of the module

Example:

>>> module.state_dict().keys()
['bias', 'weight']
to(*args, **kwargs)

Moves and/or casts the parameters and buffers.

This can be called as

to(device)
to(dtype)
to(device, dtype)

It has similar signature as torch.Tensor.to(), but does not take a Tensor and only takes in floating point dtype s. In particular, this method will only cast the floating point parameters and buffers to dtype. It will still move the integral parameters and buffers to device, if that is given. See below for examples.

Note

This method modifies the module in-place.

Parameters:
  • device (torch.device) – the desired device of the parameters and buffers in this module
  • dtype (torch.dtype) – the desired floating point type of the floating point parameters and buffers in this module
Returns:

Module – self

Example:

>>> linear = nn.Linear(2, 2)
>>> linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
>>> linear.to(torch.double)
Linear(in_features=2, out_features=2, bias=True)
>>> linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=torch.float64)
>>> gpu1 = torch.device("cuda:1")
>>> linear.to(gpu1, dtype=torch.half)
Linear(in_features=2, out_features=2, bias=True)
>>> linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=torch.float16, device='cuda:1')
>>> cpu = torch.device("cpu")
>>> linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
>>> linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=torch.float16)
train(mode=True)

Sets the module in training mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

Returns:Module – self
type(dst_type)

Casts all parameters and buffers to dst_type.

Parameters:dst_type (type or string) – the desired type
Returns:Module – self
zero_grad()

Sets gradients of all model parameters to zero.

Module contents

class mdgru.model_pytorch.mdrnn.MDGRUBlock(num_spatial_dims, dropout, num_input, num_hidden, num_output, kw)[source]

Bases: torch.nn.modules.module.Module

Convenience class combining attributes to be used for multiple MDRNN and voxel-wise fully connected layers.
param num_spatial_dims:
 Number of spatial dimensions to consider
Parameters:kw –

dict containing the following options. - resmdgru [default: False] Add a residual connection from each mdgru input to its output, possibly homogenizing dimensions using one 1 conv layer - vwfc_activation [default: class ‘torch.nn.modules.activation.Tanh’] - noactivation [default: False] - name [default: None] :param dropout: Dropout rate provided as “keep” rate :param num_input: Nuber of input units/channels :param num_hidden: number of hidden units, output units of the MDRNN :param num_output: number of output units of the voxel-wise fully connected layer

(Can be None -> no voxel-wise fully connected layer)
_all_buffers(memo=None)
_apply(fn)
_defaults = {'name': None, 'noactivation': False, 'resmdgru': {'value': False, 'help': 'Add a residual connection from each mdgru input to its output, possibly homogenizing dimensions using one 1 conv layer'}, 'vwfc_activation': <class 'torch.nn.modules.activation.Tanh'>}
_get_name()
_load_from_state_dict(state_dict, prefix, strict, missing_keys, unexpected_keys, error_msgs)

Copies parameters and buffers from state_dict into only this module, but not its descendants. This is called on every submodule in load_state_dict(). Metadata saved for this module in input state_dict is at state_dict._metadata[prefix]. Subclasses can achieve class-specific backward compatible loading using the version number at state_dict._metadata[prefix]["version"].

Note

state_dict is not the same object as the input state_dict to load_state_dict(). So it can be modified.

Parameters:
  • state_dict (dict) – a dict containing parameters and persistent buffers.
  • prefix (str) – the prefix for parameters and buffers used in this module
  • strict (bool) – whether to strictly enforce that the keys in state_dict with prefix match the names of parameters and buffers in this module
  • missing_keys (list of str) – if strict=False, add missing keys to this list
  • unexpected_keys (list of str) – if strict=False, add unexpected keys to this list
  • error_msgs (list of str) – error messages should be added to this list, and will be reported together in load_state_dict()
_slow_forward(*input, **kwargs)
_tracing_name(tracing_state)
_version = 1
add_module(name, module)

Adds a child module to the current module.

The module can be accessed as an attribute using the given name.

Parameters:
  • name (string) – name of the child module. The child module can be accessed from this module using the given name
  • parameter (Module) – child module to be added to the module.
apply(fn)

Applies fn recursively to every submodule (as returned by .children()) as well as self. Typical use includes initializing the parameters of a model (see also torch-nn-init).

Parameters:fn (Module -> None) – function to be applied to each submodule
Returns:Module – self

Example:

>>> def init_weights(m):
        print(m)
        if type(m) == nn.Linear:
            m.weight.data.fill_(1.0)
            print(m.weight)

>>> net = nn.Sequential(nn.Linear(2, 2), nn.Linear(2, 2))
>>> net.apply(init_weights)
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[ 1.,  1.],
        [ 1.,  1.]])
Linear(in_features=2, out_features=2, bias=True)
Parameter containing:
tensor([[ 1.,  1.],
        [ 1.,  1.]])
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
Sequential(
  (0): Linear(in_features=2, out_features=2, bias=True)
  (1): Linear(in_features=2, out_features=2, bias=True)
)
children()

Returns an iterator over immediate children modules.

Yields:Module – a child module
cpu()

Moves all model parameters and buffers to the CPU.

Returns:Module – self
cuda(device=None)

Moves all model parameters and buffers to the GPU.

This also makes associated parameters and buffers different objects. So it should be called before constructing optimizer if the module will live on GPU while being optimized.

Parameters:device (int, optional) – if specified, all parameters will be copied to that device
Returns:Module – self
double()

Casts all floating point parameters and buffers to double datatype.

Returns:Module – self
dump_patches = False
eval()

Sets the module in evaluation mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

extra_repr()

Set the extra representation of the module

To print customized extra information, you should reimplement this method in your own modules. Both single-line and multi-line strings are acceptable.

float()

Casts all floating point parameters and buffers to float datatype.

Returns:Module – self
forward(input)[source]
half()

Casts all floating point parameters and buffers to half datatype.

Returns:Module – self
load_state_dict(state_dict, strict=True)

Copies parameters and buffers from state_dict into this module and its descendants. If strict is True, then the keys of state_dict must exactly match the keys returned by this module’s state_dict() function.

Parameters:
  • state_dict (dict) – a dict containing parameters and persistent buffers.
  • strict (bool, optional) – whether to strictly enforce that the keys in state_dict match the keys returned by this module’s state_dict() function. Default: True
modules()

Returns an iterator over all modules in the network.

Yields:Module – a module in the network

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.modules()):
        print(idx, '->', m)

0 -> Sequential (
  (0): Linear (2 -> 2)
  (1): Linear (2 -> 2)
)
1 -> Linear (2 -> 2)
named_children()

Returns an iterator over immediate children modules, yielding both the name of the module as well as the module itself.

Yields:(string, Module) – Tuple containing a name and child module

Example:

>>> for name, module in model.named_children():
>>>     if name in ['conv4', 'conv5']:
>>>         print(module)
named_modules(memo=None, prefix='')

Returns an iterator over all modules in the network, yielding both the name of the module as well as the module itself.

Yields:(string, Module) – Tuple of name and module

Note

Duplicate modules are returned only once. In the following example, l will be returned only once.

Example:

>>> l = nn.Linear(2, 2)
>>> net = nn.Sequential(l, l)
>>> for idx, m in enumerate(net.named_modules()):
        print(idx, '->', m)

0 -> ('', Sequential (
  (0): Linear (2 -> 2)
  (1): Linear (2 -> 2)
))
1 -> ('0', Linear (2 -> 2))
named_parameters(memo=None, prefix='')

Returns an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself

Yields:(string, Parameter) – Tuple containing the name and parameter

Example:

>>> for name, param in self.named_parameters():
>>>    if name in ['bias']:
>>>        print(param.size())
parameters()

Returns an iterator over module parameters.

This is typically passed to an optimizer.

Yields:Parameter – module parameter

Example:

>>> for param in model.parameters():
>>>     print(type(param.data), param.size())
<class 'torch.FloatTensor'> (20L,)
<class 'torch.FloatTensor'> (20L, 1L, 5L, 5L)
register_backward_hook(hook)

Registers a backward hook on the module.

The hook will be called every time the gradients with respect to module inputs are computed. The hook should have the following signature:

hook(module, grad_input, grad_output) -> Tensor or None

The grad_input and grad_output may be tuples if the module has multiple inputs or outputs. The hook should not modify its arguments, but it can optionally return a new gradient with respect to input that will be used in place of grad_input in subsequent computations.

Returns:torch.utils.hooks.RemovableHandle – a handle that can be used to remove the added hook by calling handle.remove()
register_buffer(name, tensor)

Adds a persistent buffer to the module.

This is typically used to register a buffer that should not to be considered a model parameter. For example, BatchNorm’s running_mean is not a parameter, but is part of the persistent state.

Buffers can be accessed as attributes using given names.

Parameters:
  • name (string) – name of the buffer. The buffer can be accessed from this module using the given name
  • tensor (Tensor) – buffer to be registered.

Example:

>>> self.register_buffer('running_mean', torch.zeros(num_features))
register_forward_hook(hook)

Registers a forward hook on the module.

The hook will be called every time after forward() has computed an output. It should have the following signature:

hook(module, input, output) -> None

The hook should not modify the input or output.

Returns:torch.utils.hooks.RemovableHandle – a handle that can be used to remove the added hook by calling handle.remove()
register_forward_pre_hook(hook)

Registers a forward pre-hook on the module.

The hook will be called every time before forward() is invoked. It should have the following signature:

hook(module, input) -> None

The hook should not modify the input.

Returns:torch.utils.hooks.RemovableHandle – a handle that can be used to remove the added hook by calling handle.remove()
register_parameter(name, param)

Adds a parameter to the module.

The parameter can be accessed as an attribute using given name.

Parameters:
  • name (string) – name of the parameter. The parameter can be accessed from this module using the given name
  • parameter (Parameter) – parameter to be added to the module.
share_memory()
state_dict(destination=None, prefix='', keep_vars=False)

Returns a dictionary containing a whole state of the module.

Both parameters and persistent buffers (e.g. running averages) are included. Keys are corresponding parameter and buffer names.

Returns:dict – a dictionary containing a whole state of the module

Example:

>>> module.state_dict().keys()
['bias', 'weight']
to(*args, **kwargs)

Moves and/or casts the parameters and buffers.

This can be called as

to(device)
to(dtype)
to(device, dtype)

It has similar signature as torch.Tensor.to(), but does not take a Tensor and only takes in floating point dtype s. In particular, this method will only cast the floating point parameters and buffers to dtype. It will still move the integral parameters and buffers to device, if that is given. See below for examples.

Note

This method modifies the module in-place.

Parameters:
  • device (torch.device) – the desired device of the parameters and buffers in this module
  • dtype (torch.dtype) – the desired floating point type of the floating point parameters and buffers in this module
Returns:

Module – self

Example:

>>> linear = nn.Linear(2, 2)
>>> linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]])
>>> linear.to(torch.double)
Linear(in_features=2, out_features=2, bias=True)
>>> linear.weight
Parameter containing:
tensor([[ 0.1913, -0.3420],
        [-0.5113, -0.2325]], dtype=torch.float64)
>>> gpu1 = torch.device("cuda:1")
>>> linear.to(gpu1, dtype=torch.half)
Linear(in_features=2, out_features=2, bias=True)
>>> linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=torch.float16, device='cuda:1')
>>> cpu = torch.device("cpu")
>>> linear.to(cpu)
Linear(in_features=2, out_features=2, bias=True)
>>> linear.weight
Parameter containing:
tensor([[ 0.1914, -0.3420],
        [-0.5112, -0.2324]], dtype=torch.float16)
train(mode=True)

Sets the module in training mode.

This has any effect only on certain modules. See documentations of particular modules for details of their behaviors in training/evaluation mode, if they are affected, e.g. Dropout, BatchNorm, etc.

Returns:Module – self
type(dst_type)

Casts all parameters and buffers to dst_type.

Parameters:dst_type (type or string) – the desired type
Returns:Module – self
zero_grad()

Sets gradients of all model parameters to zero.