dgl | Python package built to ease deep learning | Machine Learning library

import dgl
ds = dgl.data.CSVDataset('/path/to/dataset')

# A demonstration of how to use the loaded dataset. The feature names
# may vary depending on the CSV contents.
g = ds[0] # get the graph
label = g.ndata['label']
feat = g.ndata['feat']


.. co  

在介绍DGL中邻居采样的用法之前，这里先解释一下邻居采样的工作原理。下文继续使用上述的例子。
首先定义一个如上图所示的DGLGraph。

.. code:: python

    import torch
    import dgl

    src = torch.LongTensor(
        [0, 0, 0, 1, 2, 2, 2, 3, 3, 4, 4, 5, 5, 6, 7, 7, 8, 9, 10,
         1, 2, 3, 3, 3, 4, 5, 5,   

You can use the
:ref:`same neighborhood samplers as node classification `.

.. code:: python

    sampler = dgl.dataloading.MultiLayerFullNeighborSampler(2)

To use the neighborhood sampler provided by DGL for edge classification,
one need to instea  

import dgl
import numpy as np
from heapq import heappush, heappop, heapify, nsmallest
from copy import deepcopy

# We use lapjv implementation (https://github.com/src-d/lapjv) to solve assignment problem, because of its scalability
# Also see https:/  

import argparse
import datetime
import os
import sys
import time

import dgl
import torch
from dgl.data.utils import load_graphs, save_graphs
from dgl.dataloading import GraphDataLoader
from ogb.linkproppred import DglLinkPropPredDataset, Evaluator
f  

import argparse
import math
import os
import random
import sys
import time

import numpy as np
import torch
import torch.nn.functional as F
from ogb.linkproppred import DglLinkPropPredDataset, Evaluator
from scipy.sparse.csgraph import shortest_path
  

u(t) = A*exp(t)+B*exp(-t)

v(t) = (A*exp(t)-B*exp(-t))/(A*exp(t)+B*exp(-t))

A = D*exp(-C), B = D*exp(C) ==> v(t) = tanh(t-C)

A = D*exp(-C), B =-D*exp(C) ==> v(t) = ctgh(t-C)

ctgh(t-C

import networkx as nx
import dgl

G = nx.DiGraph()
G.add_edges_from([(0, 1), (1, 2), (2, 3)])
G.nodes[0]["weight"] = 0
G.nodes[1]["weight"] = 10
G.nodes[2]["weight"] = 20
G.nodes[3]["weight"] = 30

dgl.from_networkx(G, node_attrs=["weight"

import numpy as np
import dgl
import networkx as nx

def numpy_to_graph(A,type_graph='dgl',node_features=None):
    '''Convert numpy arrays to graph

    Parameters
    ----------
    A : mxm array
        Adjacency matrix
    type_graph :

class summation: 
    def __init__(self, f, s): 
        self.first = f 
        self.second = s 
    @property
    def summ(self):
        return self.first+self.second

s = summation(1,9)
print(s.summ)
# 10
s.firs