def __init__(self,
in_dim,
out_dim,
activation,
dim,
kernel,
aggr_type,
dropout,
graph_norm,
batch_norm,
residual=False,
bias=True):
super().__init__()
self.in_dim = in_dim
self.out_dim = out_dim
self.dim = dim
self.kernel = kernel
self.graph_norm = graph_norm
self.batch_norm = batch_norm
self.residual = residual
self.dropout = dropout
self.activation = activation
self.layer = GMMConv(in_dim, out_dim, dim, kernel, aggr_type)
self.bn_node_h = nn.BatchNorm1d(out_dim)
if in_dim != out_dim:
self.residual = False
if bias:
self.bias = nn.Parameter(torch.Tensor(out_dim))
else:
self.register_buffer('bias', None)
self.reset_parameters()
def __init__(self, n_kernels, in_feats, hiddens, out_feats):
super(MoNet, self).__init__()
self.pool = nn.MaxPool1d(2)
self.layers = nn.ModuleList()
self.readout = MaxPooling()
# Input layer
self.layers.append(GMMConv(in_feats, hiddens[0], 2, n_kernels))
# Hidden layer
for i in range(1, len(hiddens)):
self.layers.append(
GMMConv(hiddens[i - 1], hiddens[i], 2, n_kernels))
self.cls = nn.Sequential(nn.Linear(hiddens[-1], out_feats),
nn.LogSoftmax())
def __init__(self, g, in_feats, n_hidden, out_feats, n_layers, dim,
n_kernels, dropout):
super(MoNet, self).__init__()
self.g = g
self.layers = nn.ModuleList()
self.pseudo_proj = nn.ModuleList()
# Input layer
self.layers.append(GMMConv(in_feats, n_hidden, dim, n_kernels))
self.pseudo_proj.append(nn.Sequential(nn.Linear(2, dim), nn.Tanh()))
# Hidden layer
for _ in range(n_layers - 1):
self.layers.append(GMMConv(n_hidden, n_hidden, dim, n_kernels))
self.pseudo_proj.append(nn.Sequential(nn.Linear(2, dim),
nn.Tanh()))
# Output layer
self.layers.append(GMMConv(n_hidden, out_feats, dim, n_kernels))
self.pseudo_proj.append(nn.Sequential(nn.Linear(2, dim), nn.Tanh()))
self.dropout = nn.Dropout(dropout)