l2gv2.embedding.gae.layers package

l2gv2.embedding.gae.layers package#

Submodules#

l2gv2.embedding.gae.layers.GAEconv module#

TODO: module docstring for embedding/gae/layers/GAEconv.py.

class l2gv2.embedding.gae.layers.GAEconv.GAEconv(dim, num_node_features, hidden_dim=32, cached=True, bias=True, add_self_loops=True, normalize=True)#

Bases: Module

implements the convolution operator for use with tg.nn.GAE

forward(data)#

compute coordinates given data

l2gv2.embedding.gae.layers.VGAEconv module#

TODO: module docstring for embedding/gae/layers/VGAEconv.py.

class l2gv2.embedding.gae.layers.VGAEconv.VGAEconv(dim, num_node_features, hidden_dim=32, cached=True, bias=True, add_self_loops=True, normalize=True)#

Bases: Module

implements the convolution operator for use with torch_geometric.nn.VGAE

forward(data: Data)#

compute mean and variance given data

Parameters:

[type] (data) – input data

Returns:

mu, sigma

l2gv2.embedding.gae.layers.decoders module#

TODO: module docstring for embedding/gae/layers/decoders.py.

class l2gv2.embedding.gae.layers.decoders.DistanceDecoder#

Bases: Module

implements the distance decoder which predicts the probability of an edge as the exponential of the negative euclidean distance between nodes

forward(z, edge_index, sigmoid=True)#

compute decoder values

Parameters:

  • [type] (sigmoid) – input coordinates

  • [type] – edges

  • [type] – if True, return exponential of negative distance, else return negative distance

forward_all(z, sigmoid=True)#

compute value for all node pairs

Parameters:

  • [type] (sigmoid) – input coordinates

  • [type] – if True, return exponential of negative distance, else return negative distance

Module contents#

TODO: module docstring for embedding/gae/layers/__init__.py.

class l2gv2.embedding.gae.layers.DistanceDecoder#

Bases: Module

implements the distance decoder which predicts the probability of an edge as the exponential of the negative euclidean distance between nodes

forward(z, edge_index, sigmoid=True)#

compute decoder values

Parameters:

  • [type] (sigmoid) – input coordinates

  • [type] – edges

  • [type] – if True, return exponential of negative distance, else return negative distance

forward_all(z, sigmoid=True)#

compute value for all node pairs

Parameters:

  • [type] (sigmoid) – input coordinates

  • [type] – if True, return exponential of negative distance, else return negative distance

class l2gv2.embedding.gae.layers.GAEconv(dim, num_node_features, hidden_dim=32, cached=True, bias=True, add_self_loops=True, normalize=True)#

Bases: Module

implements the convolution operator for use with tg.nn.GAE

forward(data)#

compute coordinates given data

class l2gv2.embedding.gae.layers.VGAEconv(dim, num_node_features, hidden_dim=32, cached=True, bias=True, add_self_loops=True, normalize=True)#

Bases: Module

implements the convolution operator for use with torch_geometric.nn.VGAE

forward(data: Data)#

compute mean and variance given data

Parameters:

[type] (data) – input data

Returns:

mu, sigma

Contents