Basic networkx



In [1]:

    
import networkx as nx



In [2]:

    
G = nx.DiGraph() # directed Graph



In [3]:

    
G.add_edge( 1,2 )  # nodes will be automatically created from edges
G.nodes()









    Out[3]:





[1, 2]



In [4]:

    
myedges = [ (2,1), (3,1), (1,3) ] # a list of tuples
G.add_edges_from( myedges ) # ading edges from a list of tuples
G.nodes()









    Out[4]:





[1, 2, 3]

Plotting the Graph



In [13]:

    
pos = nx.spring_layout(G) # positions for all node
nx.draw(G, pos, node_color = '#A0CBE2', width=1, with_labels=1)

Edges Properties

When calling nodes, we obtain the edges and its properties



In [5]:

    
G[1] # node 1 is connected to 2 and 3









    Out[5]:





{2: {}, 3: {}}



In [14]:

    
G.neighbors(1) # to check only connected nodes









    Out[14]:





[2, 3]



In [6]:

    
G[1][3]['prob'] = 0.25 # setting a 'probability' item to edge 1-3
G[1]









    Out[6]:





{2: {}, 3: {'prob': 0.25}}



In [7]:

    
G[1][3].items() # see their edges properties as a list of tuples









    Out[7]:





[('prob', 0.25)]



In [15]:

    
G.add_weighted_edges_from([ (3, 4, 0.5) ]) # setting weight and creating a new node (4)
G[3]









    Out[15]:





{1: {}, 4: {'weight': 0.5}}

Iterations



In [17]:

    
H = nx.DiGraph() # create a new graph
H.add_edges_from(G.edges(), weigth=0.5) # use edges of G to create H with weigth 0.5



In [18]:

    
H.edges()









    Out[18]:





[(1, 2), (1, 3), (2, 1), (3, 1), (3, 4)]

Fast iteration of edges with adjacent interators with .adjacent_iter()



In [21]:

    
for node, neighbor in H.adjacency_iter(): # adjacentcy_iter return a dictionary with the  edges
    print(node, neighbor) # dictionary of dictionaries, get









    



(1, {2: {'weigth': 0.5}, 3: {'weigth': 0.5}})
(2, {1: {'weigth': 0.5}})
(3, {1: {'weigth': 0.5}, 4: {'weigth': 0.5}})
(4, {})

List of attributes of the Graph



In [22]:

    
for node, neighbor in H.adjacency_iter():
    for n, myattdict in neighbor.items():
        print( myattdict['weigth'] )

List all edeges and its attributes



In [23]:

    
for node, neighbor in H.adjacency_iter():
    for n, myattdict in neighbor.items():
        if myattdict['weigth'] < 0.6: 
            print(node, n, myattdict['weigth'])









    



(1, 2, 0.5)
(1, 3, 0.5)
(2, 1, 0.5)
(3, 1, 0.5)
(3, 4, 0.5)



In [ ]: