cogdl.layers.prone_module

Module Contents

Classes

HeatKernel

HeatKernelApproximation

Gaussian

PPR

applying sparsification to accelerate computation

SignalRescaling

  • rescale signal of each node according to the degree of the node:

ProNE

NodeAdaptiveEncoder

  • shrink negative values in signal/feature matrix

Functions

propagate(mx, emb, stype, space=None)

get_embedding_dense(matrix, dimension)

class cogdl.layers.prone_module.HeatKernel(t=0.5, theta0=0.6, theta1=0.4)[source]

Bases: object

prop_adjacency(self, mx)[source]
prop(self, mx, emb)[source]
class cogdl.layers.prone_module.HeatKernelApproximation(t=0.2, k=5)[source]

Bases: object

taylor(self, mx, emb)[source]
chebyshev(self, mx, emb)[source]
prop(self, mx, emb)[source]
class cogdl.layers.prone_module.Gaussian(mu=0.5, theta=1, rescale=False, k=3)[source]

Bases: object

prop(self, mx, emb)[source]
class cogdl.layers.prone_module.PPR(alpha=0.5, k=10)[source]

Bases: object

applying sparsification to accelerate computation

prop(self, mx, emb)[source]
class cogdl.layers.prone_module.SignalRescaling[source]

Bases: object

  • rescale signal of each node according to the degree of the node:

  • sigmoid(degree)

  • sigmoid(1/degree)

prop(self, mx, emb)[source]
class cogdl.layers.prone_module.ProNE[source]

Bases: object

__call__(self, A, a, order=10, mu=0.1, s=0.5)[source]
class cogdl.layers.prone_module.NodeAdaptiveEncoder[source]

Bases: object

  • shrink negative values in signal/feature matrix

  • no learning

static prop(signal)[source]
cogdl.layers.prone_module.propagate(mx, emb, stype, space=None)[source]
cogdl.layers.prone_module.get_embedding_dense(matrix, dimension)[source]