dmlc--dgl
Zihao Ye
9a0511c8e9
* upd * damn it * fuck * fuck pylint * fudge * remove some comments about MXNet * upd * upd * damn it * damn it * fuck * fuck * upd * upd * pylint bastard * upd * upd * upd * upd * upd * upd * upd * upd * upd
56 行
1.6 KiB
Python
56 行
1.6 KiB
Python
"""
|
|
APPNP implementation in DGL.
|
|
References
|
|
----------
|
|
Paper: https://arxiv.org/abs/1810.05997
|
|
Author's code: https://github.com/klicperajo/ppnp
|
|
"""
|
|
import torch.nn as nn
|
|
from dgl.nn.pytorch.conv import APPNPConv
|
|
|
|
|
|
class APPNP(nn.Module):
|
|
def __init__(self,
|
|
g,
|
|
in_feats,
|
|
hiddens,
|
|
n_classes,
|
|
activation,
|
|
feat_drop,
|
|
edge_drop,
|
|
alpha,
|
|
k):
|
|
super(APPNP, self).__init__()
|
|
self.g = g
|
|
self.layers = nn.ModuleList()
|
|
# input layer
|
|
self.layers.append(nn.Linear(in_feats, hiddens[0]))
|
|
# hidden layers
|
|
for i in range(1, len(hiddens)):
|
|
self.layers.append(nn.Linear(hiddens[i - 1], hiddens[i]))
|
|
# output layer
|
|
self.layers.append(nn.Linear(hiddens[-1], n_classes))
|
|
self.activation = activation
|
|
if feat_drop:
|
|
self.feat_drop = nn.Dropout(feat_drop)
|
|
else:
|
|
self.feat_drop = lambda x: x
|
|
self.propagate = APPNPConv(k, alpha, edge_drop)
|
|
self.reset_parameters()
|
|
|
|
def reset_parameters(self):
|
|
for layer in self.layers:
|
|
layer.reset_parameters()
|
|
|
|
def forward(self, features):
|
|
# prediction step
|
|
h = features
|
|
h = self.feat_drop(h)
|
|
h = self.activation(self.layers[0](h))
|
|
for layer in self.layers[1:-1]:
|
|
h = self.activation(layer(h))
|
|
h = self.layers[-1](self.feat_drop(h))
|
|
# propagation step
|
|
h = self.propagate(self.g, h)
|
|
return h
|