models
¶
Subpackages¶
models.emb
models.emb.deepwalk
models.emb.dgk
models.emb.dngr
models.emb.gatne
models.emb.graph2vec
models.emb.grarep
models.emb.hin2vec
models.emb.hope
models.emb.line
models.emb.metapath2vec
models.emb.netmf
models.emb.netsmf
models.emb.node2vec
models.emb.prone
models.emb.pte
models.emb.sdne
models.emb.spectral
models.nn
models.nn.asgcn
models.nn.compgcn
models.nn.dgi
models.nn.dgl_gcc
models.nn.disengcn
models.nn.fastgcn
models.nn.gat
models.nn.gcn
models.nn.gcnmix
models.nn.grand
models.nn.graphsage
models.nn.mixhop
models.nn.mlp
models.nn.mvgrl
models.nn.patchy_san
models.nn.pyg_cheb
models.nn.pyg_dgcnn
models.nn.pyg_diffpool
models.nn.pyg_drgat
models.nn.pyg_drgcn
models.nn.pyg_gat
models.nn.pyg_gcn
models.nn.pyg_gin
models.nn.pyg_gtn
models.nn.pyg_han
models.nn.pyg_infograph
models.nn.pyg_infomax
models.nn.pyg_sortpool
models.nn.pyg_srgcn
models.nn.pyg_unet
models.nn.rgcn
models.nn.unsup_graphsage
Submodules¶
Package Contents¶
Functions¶
|
New model types can be added to cogdl with the |
|
Compute utility lists for non-uniform sampling from discrete distributions. |
|
Draw sample from a non-uniform discrete distribution using alias sampling. |
|
-
class
models.
BaseModel
[source]¶ Bases:
torch.nn.Module
-
static
add_args
(parser)¶ Add model-specific arguments to the parser.
-
abstract classmethod
build_model_from_args
(cls, args)¶ Build a new model instance.
-
static
-
models.
register_model
(name)[source]¶ New model types can be added to cogdl with the
register_model()
function decorator.For example:
@register_model('gat') class GAT(BaseModel): (...)
- Args:
name (str): the name of the model
-
models.
alias_setup
(probs)[source]¶ Compute utility lists for non-uniform sampling from discrete distributions. Refer to https://hips.seas.harvard.edu/blog/2013/03/03/the-alias-method-efficient-sampling-with-many-discrete-outcomes/ for details