AllSet_Transformer_Layer

AllSetTransformer Layer Module.

class topomodelx.nn.hypergraph.allset_transformer_layer.AllSetTransformerBlock(in_channels, hidden_channels, heads: int = 4, number_queries: int = 1, dropout: float = 0.0, mlp_num_layers: int = 1, mlp_activation=<class 'torch.nn.modules.activation.ReLU'>, mlp_dropout: float = 0.0, mlp_norm=None, initialization: Literal['xavier_uniform', 'xavier_normal']='xavier_uniform')

AllSetTransformer Block Module.

A module for AllSet Transformer block in a bipartite graph.

Parameters:

in_channelsint

Dimension of the input features.

hidden_channelsint

Dimension of the hidden features.

headsint, default=4

Number of attention heads.

number_queriesint, default=1

Number of queries.

dropoutfloat, default=0.0

Dropout probability.

mlp_num_layersint, default=1

Number of layers in the MLP.

mlp_activationcallable or None, optional

Activation function in the MLP.

mlp_dropoutfloat, optional

Dropout probability in the MLP.

mlp_normstr or None, optional

Type of layer normalization in the MLP.

initializationLiteral[“xavier_uniform”, “xavier_normal”], default=”xavier_uniform”

Initialization method.

Methods

add_module(name, module)	Add a child module to the current module.
apply(fn)	Apply fn recursively to every submodule (as returned by .children()) as well as self.
bfloat16()	Casts all floating point parameters and buffers to bfloat16 datatype.
buffers([recurse])	Return an iterator over module buffers.
children()	Return an iterator over immediate children modules.
compile(*args, **kwargs)	Compile this Module's forward using torch.compile().
cpu()	Move all model parameters and buffers to the CPU.
cuda([device])	Move all model parameters and buffers to the GPU.
double()	Casts all floating point parameters and buffers to double datatype.
eval()	Set the module in evaluation mode.
extra_repr()	Return the extra representation of the module.
float()	Casts all floating point parameters and buffers to float datatype.
forward(x_source, neighborhood)	Forward computation.
get_buffer(target)	Return the buffer given by target if it exists, otherwise throw an error.
get_extra_state()	Return any extra state to include in the module's state_dict.
get_parameter(target)	Return the parameter given by target if it exists, otherwise throw an error.
get_submodule(target)	Return the submodule given by target if it exists, otherwise throw an error.
half()	Casts all floating point parameters and buffers to half datatype.
ipu([device])	Move all model parameters and buffers to the IPU.
load_state_dict(state_dict[, strict, assign])	Copy parameters and buffers from state_dict into this module and its descendants.
modules([remove_duplicate])	Return an iterator over all modules in the network.
mtia([device])	Move all model parameters and buffers to the MTIA.
named_buffers([prefix, recurse, ...])	Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.
named_children()	Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.
named_modules([memo, prefix, remove_duplicate])	Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.
named_parameters([prefix, recurse, ...])	Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.
parameters([recurse])	Return an iterator over module parameters.
register_backward_hook(hook)	Register a backward hook on the module.
register_buffer(name, tensor[, persistent])	Add a buffer to the module.
register_forward_hook(hook, *[, prepend, ...])	Register a forward hook on the module.
register_forward_pre_hook(hook, *[, ...])	Register a forward pre-hook on the module.
register_full_backward_hook(hook[, prepend])	Register a backward hook on the module.
register_full_backward_pre_hook(hook[, prepend])	Register a backward pre-hook on the module.
register_load_state_dict_post_hook(hook)	Register a post-hook to be run after module's load_state_dict() is called.
register_load_state_dict_pre_hook(hook)	Register a pre-hook to be run before module's load_state_dict() is called.
register_module(name, module)	Alias for add_module().
register_parameter(name, param)	Add a parameter to the module.
register_state_dict_post_hook(hook)	Register a post-hook for the state_dict() method.
register_state_dict_pre_hook(hook)	Register a pre-hook for the state_dict() method.
requires_grad_([requires_grad])	Change if autograd should record operations on parameters in this module.
reset_parameters()	Reset learnable parameters.
set_extra_state(state)	Set extra state contained in the loaded state_dict.
set_submodule(target, module[, strict])	Set the submodule given by target if it exists, otherwise throw an error.
share_memory()	See torch.Tensor.share_memory_().
state_dict(*args[, destination, prefix, ...])	Return a dictionary containing references to the whole state of the module.
to(*args, **kwargs)	Move and/or cast the parameters and buffers.
to_empty(*, device[, recurse])	Move the parameters and buffers to the specified device without copying storage.
train([mode])	Set the module in training mode.
type(dst_type)	Casts all parameters and buffers to dst_type.
xpu([device])	Move all model parameters and buffers to the XPU.
zero_grad([set_to_none])	Reset gradients of all model parameters.

__call__

forward(x_source, neighborhood)

Forward computation.

Parameters:

x_sourceTensor, shape = (…, n_source_cells, in_channels)

Inputer features.

neighborhoodtorch.sparse, shape = (n_target_cells, n_source_cells)

Neighborhood matrix.

Returns:

x_message_on_target

Output sum over features on target cells.

reset_parameters() → None

Reset learnable parameters.

class topomodelx.nn.hypergraph.allset_transformer_layer.AllSetTransformerLayer(in_channels, hidden_channels, heads: int = 4, number_queries: int = 1, dropout: float = 0.0, mlp_num_layers: int = 1, mlp_activation=<class 'torch.nn.modules.activation.ReLU'>, mlp_dropout: float = 0.0, mlp_norm=None, **kwargs)

Implementation of the AllSetTransformer Layer proposed in [1].

Parameters:

in_channelsint

Dimension of the input features.

hidden_channelsint

Dimension of the hidden features.

headsint, default=4

Number of attention heads.

number_queriesint, default=1

Number of queries.

dropoutfloat, optional

Dropout probability.

mlp_num_layersint, default=1

Number of layers in the MLP.

mlp_activationcallable or None, optional

Activation function in the MLP.

mlp_dropoutfloat, optional

Dropout probability in the MLP.

mlp_normstr or None, optional

Type of layer normalization in the MLP.

**kwargsoptional

Additional arguments for the layer modules.

Methods

add_module(name, module)	Add a child module to the current module.
apply(fn)	Apply fn recursively to every submodule (as returned by .children()) as well as self.
bfloat16()	Casts all floating point parameters and buffers to bfloat16 datatype.
buffers([recurse])	Return an iterator over module buffers.
children()	Return an iterator over immediate children modules.
compile(*args, **kwargs)	Compile this Module's forward using torch.compile().
cpu()	Move all model parameters and buffers to the CPU.
cuda([device])	Move all model parameters and buffers to the GPU.
double()	Casts all floating point parameters and buffers to double datatype.
eval()	Set the module in evaluation mode.
extra_repr()	Return the extra representation of the module.
float()	Casts all floating point parameters and buffers to float datatype.
forward(x_0, incidence_1)	Forward computation.
get_buffer(target)	Return the buffer given by target if it exists, otherwise throw an error.
get_extra_state()	Return any extra state to include in the module's state_dict.
get_parameter(target)	Return the parameter given by target if it exists, otherwise throw an error.
get_submodule(target)	Return the submodule given by target if it exists, otherwise throw an error.
half()	Casts all floating point parameters and buffers to half datatype.
ipu([device])	Move all model parameters and buffers to the IPU.
load_state_dict(state_dict[, strict, assign])	Copy parameters and buffers from state_dict into this module and its descendants.
modules([remove_duplicate])	Return an iterator over all modules in the network.
mtia([device])	Move all model parameters and buffers to the MTIA.
named_buffers([prefix, recurse, ...])	Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.
named_children()	Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.
named_modules([memo, prefix, remove_duplicate])	Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.
named_parameters([prefix, recurse, ...])	Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.
parameters([recurse])	Return an iterator over module parameters.
register_backward_hook(hook)	Register a backward hook on the module.
register_buffer(name, tensor[, persistent])	Add a buffer to the module.
register_forward_hook(hook, *[, prepend, ...])	Register a forward hook on the module.
register_forward_pre_hook(hook, *[, ...])	Register a forward pre-hook on the module.
register_full_backward_hook(hook[, prepend])	Register a backward hook on the module.
register_full_backward_pre_hook(hook[, prepend])	Register a backward pre-hook on the module.
register_load_state_dict_post_hook(hook)	Register a post-hook to be run after module's load_state_dict() is called.
register_load_state_dict_pre_hook(hook)	Register a pre-hook to be run before module's load_state_dict() is called.
register_module(name, module)	Alias for add_module().
register_parameter(name, param)	Add a parameter to the module.
register_state_dict_post_hook(hook)	Register a post-hook for the state_dict() method.
register_state_dict_pre_hook(hook)	Register a pre-hook for the state_dict() method.
requires_grad_([requires_grad])	Change if autograd should record operations on parameters in this module.
reset_parameters()	Reset parameters.
set_extra_state(state)	Set extra state contained in the loaded state_dict.
set_submodule(target, module[, strict])	Set the submodule given by target if it exists, otherwise throw an error.
share_memory()	See torch.Tensor.share_memory_().
state_dict(*args[, destination, prefix, ...])	Return a dictionary containing references to the whole state of the module.
to(*args, **kwargs)	Move and/or cast the parameters and buffers.
to_empty(*, device[, recurse])	Move the parameters and buffers to the specified device without copying storage.
train([mode])	Set the module in training mode.
type(dst_type)	Casts all parameters and buffers to dst_type.
xpu([device])	Move all model parameters and buffers to the XPU.
zero_grad([set_to_none])	Reset gradients of all model parameters.

__call__

References

[1]

Chien, Pan, Peng and Milenkovic. You are AllSet: a multiset function framework for hypergraph neural networks. ICLR 2022. https://arxiv.org/abs/2106.13264

forward(x_0, incidence_1)

Forward computation.

Vertex to edge:

\[\begin{split}\begin{align*} &🟧 \quad m_{\rightarrow z}^{(\rightarrow 1)} = AGG_{y \in \\mathcal{B}(z)} (h_y^{t, (0)}, h_z^{t,(1)}) \quad \text{with attention}\\ &🟦 \quad h_z^{t+1,(1)} = \text{LN}(m_{\rightarrow z}^{(\rightarrow 1)} + \text{MLP}(m_{\rightarrow z}^{(\rightarrow 1)} )) \end{align*}\end{split}\]

Edge to vertex:

\[\begin{split}\begin{align*} &🟧 \quad m_{\rightarrow x}^{(\rightarrow 0)} = AGG_{z \in \mathcal{C}(x)} (h_z^{t+1,(1)}, h_x^{t,(0)}) \quad \text{with attention}\\ &🟦 \quad h_x^{t+1,(0)} = \text{LN}(m_{\rightarrow x}^{(\rightarrow 0)} + \text{MLP}(m_{\rightarrow x}^{(\rightarrow 0)} )) \end{align*}\end{split}\]

Parameters:

x_0torch.Tensor, shape = (n_nodes, channels)

Node input features.

incidence_1torch.sparse, shape = (n_nodes, n_hyperedges)

Incidence matrix \(B_1\) mapping hyperedges to nodes.

Returns:

x_0torch.Tensor

Output node features.

x_1torch.Tensor

Output hyperedge features.

reset_parameters() → None

Reset parameters.

class topomodelx.nn.hypergraph.allset_transformer_layer.MLP(in_channels, hidden_channels, norm_layer=None, activation_layer=<class 'torch.nn.modules.activation.ReLU'>, dropout: float = 0.0, inplace: bool | None = None, bias: bool = False)

MLP Module.

A module for a multi-layer perceptron (MLP).

Parameters:

in_channelsint

Dimension of the input features.

hidden_channelslist of int

List of dimensions of the hidden features.

norm_layercallable or None, optional

Type of layer normalization.

activation_layercallable or None, optional

Type of activation function.

dropoutfloat, optional

Dropout probability.

inplacebool, default=False

Whether to do the operation in-place.

biasbool, default=False

Whether to add bias.

Methods

add_module(name, module)	Add a child module to the current module.
append(module)	Append a given module to the end.
apply(fn)	Apply fn recursively to every submodule (as returned by .children()) as well as self.
bfloat16()	Casts all floating point parameters and buffers to bfloat16 datatype.
buffers([recurse])	Return an iterator over module buffers.
children()	Return an iterator over immediate children modules.
compile(*args, **kwargs)	Compile this Module's forward using torch.compile().
cpu()	Move all model parameters and buffers to the CPU.
cuda([device])	Move all model parameters and buffers to the GPU.
double()	Casts all floating point parameters and buffers to double datatype.
eval()	Set the module in evaluation mode.
extend(sequential)	Extends the current Sequential container with layers from another Sequential container.
extra_repr()	Return the extra representation of the module.
float()	Casts all floating point parameters and buffers to float datatype.
forward(input)	Runs the forward pass.
get_buffer(target)	Return the buffer given by target if it exists, otherwise throw an error.
get_extra_state()	Return any extra state to include in the module's state_dict.
get_parameter(target)	Return the parameter given by target if it exists, otherwise throw an error.
get_submodule(target)	Return the submodule given by target if it exists, otherwise throw an error.
half()	Casts all floating point parameters and buffers to half datatype.
insert(index, module)	Inserts a module into the Sequential container at the specified index.
ipu([device])	Move all model parameters and buffers to the IPU.
load_state_dict(state_dict[, strict, assign])	Copy parameters and buffers from state_dict into this module and its descendants.
modules([remove_duplicate])	Return an iterator over all modules in the network.
mtia([device])	Move all model parameters and buffers to the MTIA.
named_buffers([prefix, recurse, ...])	Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.
named_children()	Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.
named_modules([memo, prefix, remove_duplicate])	Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.
named_parameters([prefix, recurse, ...])	Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.
parameters([recurse])	Return an iterator over module parameters.
pop(key)	Pop key from self.
register_backward_hook(hook)	Register a backward hook on the module.
register_buffer(name, tensor[, persistent])	Add a buffer to the module.
register_forward_hook(hook, *[, prepend, ...])	Register a forward hook on the module.
register_forward_pre_hook(hook, *[, ...])	Register a forward pre-hook on the module.
register_full_backward_hook(hook[, prepend])	Register a backward hook on the module.
register_full_backward_pre_hook(hook[, prepend])	Register a backward pre-hook on the module.
register_load_state_dict_post_hook(hook)	Register a post-hook to be run after module's load_state_dict() is called.
register_load_state_dict_pre_hook(hook)	Register a pre-hook to be run before module's load_state_dict() is called.
register_module(name, module)	Alias for add_module().
register_parameter(name, param)	Add a parameter to the module.
register_state_dict_post_hook(hook)	Register a post-hook for the state_dict() method.
register_state_dict_pre_hook(hook)	Register a pre-hook for the state_dict() method.
requires_grad_([requires_grad])	Change if autograd should record operations on parameters in this module.
set_extra_state(state)	Set extra state contained in the loaded state_dict.
set_submodule(target, module[, strict])	Set the submodule given by target if it exists, otherwise throw an error.
share_memory()	See torch.Tensor.share_memory_().
state_dict(*args[, destination, prefix, ...])	Return a dictionary containing references to the whole state of the module.
to(*args, **kwargs)	Move and/or cast the parameters and buffers.
to_empty(*, device[, recurse])	Move the parameters and buffers to the specified device without copying storage.
train([mode])	Set the module in training mode.
type(dst_type)	Casts all parameters and buffers to dst_type.
xpu([device])	Move all model parameters and buffers to the XPU.
zero_grad([set_to_none])	Reset gradients of all model parameters.

__call__

class topomodelx.nn.hypergraph.allset_transformer_layer.MultiHeadAttention(in_channels, hidden_channels, aggr_norm: bool = False, update_func=None, heads: int = 4, number_queries: int = 1, initialization: Literal['xavier_uniform', 'xavier_normal'] = 'xavier_uniform', initialization_gain: float = 1.414)

Computes the multi-head attention mechanism (QK^T)V of transformer-based architectures.

MH module from Eq(7) in AllSet paper [1]_.

Parameters:

in_channelsint

Dimension of input features.

hidden_channelsint

Dimension of hidden features.

aggr_normbool, default=False

Whether to normalize the aggregated message by the neighborhood size.

update_funcstr or None, optional

Update method to apply to message.

headsint, default=4

Number of attention heads.

number_queriesint, default=1

Number of queries.

initializationLiteral[“xavier_uniform”, “xavier_normal”], default=”xavier_uniform”

Initialization method.

initialization_gainfloat, default=1.414

Gain factor for initialization.

Methods

add_module(name, module)	Add a child module to the current module.
aggregate(x_message)	Aggregate messages on each target cell.
apply(fn)	Apply fn recursively to every submodule (as returned by .children()) as well as self.
attention(x_source, neighborhood)	Compute (QK^T) of transformer-based architectures.
bfloat16()	Casts all floating point parameters and buffers to bfloat16 datatype.
buffers([recurse])	Return an iterator over module buffers.
children()	Return an iterator over immediate children modules.
compile(*args, **kwargs)	Compile this Module's forward using torch.compile().
cpu()	Move all model parameters and buffers to the CPU.
cuda([device])	Move all model parameters and buffers to the GPU.
double()	Casts all floating point parameters and buffers to double datatype.
eval()	Set the module in evaluation mode.
extra_repr()	Return the extra representation of the module.
float()	Casts all floating point parameters and buffers to float datatype.
forward(x_source, neighborhood)	Forward pass.
get_buffer(target)	Return the buffer given by target if it exists, otherwise throw an error.
get_extra_state()	Return any extra state to include in the module's state_dict.
get_parameter(target)	Return the parameter given by target if it exists, otherwise throw an error.
get_submodule(target)	Return the submodule given by target if it exists, otherwise throw an error.
half()	Casts all floating point parameters and buffers to half datatype.
ipu([device])	Move all model parameters and buffers to the IPU.
load_state_dict(state_dict[, strict, assign])	Copy parameters and buffers from state_dict into this module and its descendants.
message(x_source[, x_target])	Construct message from source cells to target cells.
modules([remove_duplicate])	Return an iterator over all modules in the network.
mtia([device])	Move all model parameters and buffers to the MTIA.
named_buffers([prefix, recurse, ...])	Return an iterator over module buffers, yielding both the name of the buffer as well as the buffer itself.
named_children()	Return an iterator over immediate children modules, yielding both the name of the module as well as the module itself.
named_modules([memo, prefix, remove_duplicate])	Return an iterator over all modules in the network, yielding both the name of the module as well as the module itself.
named_parameters([prefix, recurse, ...])	Return an iterator over module parameters, yielding both the name of the parameter as well as the parameter itself.
parameters([recurse])	Return an iterator over module parameters.
register_backward_hook(hook)	Register a backward hook on the module.
register_buffer(name, tensor[, persistent])	Add a buffer to the module.
register_forward_hook(hook, *[, prepend, ...])	Register a forward hook on the module.
register_forward_pre_hook(hook, *[, ...])	Register a forward pre-hook on the module.
register_full_backward_hook(hook[, prepend])	Register a backward hook on the module.
register_full_backward_pre_hook(hook[, prepend])	Register a backward pre-hook on the module.
register_load_state_dict_post_hook(hook)	Register a post-hook to be run after module's load_state_dict() is called.
register_load_state_dict_pre_hook(hook)	Register a pre-hook to be run before module's load_state_dict() is called.
register_module(name, module)	Alias for add_module().
register_parameter(name, param)	Add a parameter to the module.
register_state_dict_post_hook(hook)	Register a post-hook for the state_dict() method.
register_state_dict_pre_hook(hook)	Register a pre-hook for the state_dict() method.
requires_grad_([requires_grad])	Change if autograd should record operations on parameters in this module.
reset_parameters()	Reset learnable parameters.
set_extra_state(state)	Set extra state contained in the loaded state_dict.
set_submodule(target, module[, strict])	Set the submodule given by target if it exists, otherwise throw an error.
share_memory()	See torch.Tensor.share_memory_().
state_dict(*args[, destination, prefix, ...])	Return a dictionary containing references to the whole state of the module.
to(*args, **kwargs)	Move and/or cast the parameters and buffers.
to_empty(*, device[, recurse])	Move the parameters and buffers to the specified device without copying storage.
train([mode])	Set the module in training mode.
type(dst_type)	Casts all parameters and buffers to dst_type.
xpu([device])	Move all model parameters and buffers to the XPU.
zero_grad([set_to_none])	Reset gradients of all model parameters.

__call__

attention(x_source, neighborhood)

Compute (QK^T) of transformer-based architectures.

Parameters:

x_sourcetorch.Tensor, shape = (n_source_cells, in_channels)

Input features on source cells. Assumes that all source cells have the same rank r.

neighborhoodtorch.sparse, shape = (n_target_cells, n_source_cells)

Neighborhood matrix.

Returns:

torch.Tensor, shape = (n_target_cells, heads, number_queries, n_source_cells)

Attention weights: one scalar per message between a source and a target cell.

forward(x_source, neighborhood)

Forward pass.

Computes (QK^T)V attention mechanism of transformer-based architectures. Module MH from Eq (7) in AllSet paper [1]_.

Parameters:

x_sourceTensor, shape = (…, n_source_cells, in_channels)

Input features on source cells. Assumes that all source cells have the same rank r.

neighborhoodtorch.sparse, shape = (n_target_cells, n_source_cells)

Neighborhood matrix.

Returns:

Tensor, shape = (…, n_target_cells, out_channels)

Output features on target cells. Assumes that all target cells have the same rank s.

reset_parameters()

Reset learnable parameters.