如何评价AWS的图神经网络框架DGL？ 第1页

       import dgl import mxnet as mx  g = dgl.DGLGraph() g.add_nodes(5)                           # add 5 nodes g.add_edges([0, 0, 0, 0], [1, 2, 3, 4])  # add 4 edges 0->1, 0->2, 0->3, 0->4 g.ndata['h'] = mx.nd.random.randn(5, 3)  # assign feature  to each node g.edata['h'] = mx.nd.random.randn(4, 4)  # assign feature to each edge     

       import dgl import torch as th g = dgl.DGLGraph() g.add_nodes(5)                           # add 5 nodes g.add_edges([0, 0, 0, 0], [1, 2, 3, 4])  # add 4 edges 0->1, 0->2, 0->3, 0->4 g.ndata['h'] = th.randn(5, 3)            # assign feature  to each node g.edata['h'] = th.randn(4, 4)            # assign feature to each edge      

       import numpy as np import mxnet as mx from mxnet import gluon import dgl from dgl import DGLGraph  dgl.load_backend(os.environ.get('DGLBACKEND', 'mxnet').lower())  # set mxnet as backend  def gcn_msg(edges):  # message passing     return {'m' : edges.src['h']}  def gcn_reduce(nodes):  # sum the the feature of incoming node     return {'h' : mx.nd.sum(nodes.mailbox['m'], 1)}   class NodeUpdateModule(gluon.Block):  # define the GCN layer     def __init__(self, out_feats, activation=None, dropout=0):         super(NodeUpdateModule, self).__init__()         self.linear = gluon.nn.Dense(out_feats, activation=activation)         self.dropout = dropout      def forward(self, node):         h = self.linear(node.data['h'])         if self.dropout:             h = mx.nd.Dropout(h, p=self.dropout)         return {'h': h}   class GCN(gluon.Block):  # define the model     def __init__(self,                  g,                  in_feats,                  n_hidden,                  n_classes,                  n_layers,                  activation,                  dropout,                  ):         super(GCN, self).__init__()         self.g = g         self.dropout = dropout         self.conv_layers = gluon.nn.Sequential()         for i in range(n_layers):             self.conv_layers.add(NodeUpdateModule(n_hidden, activation, dropout))          self.out_layer = gluon.nn.Dense(n_classes)         self.gcn_msg = gcn_msg         self.gcn_reduce = gcn_reduce      def forward(self, features):         self.g.ndata['h'] = features         for layer in self.conv_layers:             self.g.update_all(self.gcn_msg, self.gcn_reduce, layer)  # update the graph         return self.out_layer(self.g.ndata.pop('h'))        

       import numpy as np import torch import torch.nn as nn import torch.nn.functional as F from dgl import DGLGraph from dgl.data import register_data_args, load_data  def gcn_msg(edges):     return {'m' : edges.src['h']}  def gcn_reduce(nodes):     return {'h' : torch.sum(nodes.mailbox['m'], 1)}  class NodeApplyModule(nn.Module):  # define the GCN layer.     def __init__(self, in_feats, out_feats, activation=None):         super(NodeApplyModule, self).__init__()         self.linear = nn.Linear(in_feats, out_feats)         self.activation = activation      def forward(self, nodes):         # normalization by square root of dst degree         h = nodes.data['h']          h = self.linear(h)         if self.activation:             h = self.activation(h)         return {'h' : h}  class GCN(nn.Module):  # define  the model      def __init__(self,                  g,                  in_feats,                  n_hidden,                  n_classes,                  n_layers,                  activation,                  dropout):         super(GCN, self).__init__()         self.g = g          if dropout:             self.dropout = nn.Dropout(p=dropout)         else:             self.dropout = 0.          self.layers = nn.ModuleList()          # input layer         self.layers.append(NodeApplyModule(in_feats, n_hidden, activation))          # hidden layers         for i in range(n_layers - 1):             self.layers.append(NodeApplyModule(n_hidden, n_hidden, activation))          # output layer         self.layers.append(NodeApplyModule(n_hidden, n_classes))      def forward(self, features):         self.g.ndata['h'] = features          for idx, layer in enumerate(self.layers):             # apply dropout             if idx > 0 and self.dropout:                 self.g.ndata['h'] = self.dropout(self.g.ndata['h'])             self.g.ndata['h'] = self.g.ndata['h']              self.g.update_all(gcn_msg, gcn_reduce, layer)         return self.g.ndata.pop('h')