A method of drawing anchored maps for bipartite graphs
the method about how to decide the order of anchor nodes from the paper Drawing Bipartite Graphs as Anchored Maps， and use Fruchterman-Reingold algorithm (this part borrows from network's spring_layout，if you want to kown more , you can see the paper Graph Drawing by Force-directed Placement) to layout the free node。
There are many parameters, you can choose tnem according to your needs，I will introduce some commonly used parameter here。
In this case,you just need to specify the source data path, don't need to modify other parameters .But the algorithm has certain requirements for the format of the data. You can specify whether it is a member or a community through the location attribute of the node. The algorithm will treat community node as an anchor.
json_file = './test.json'
pos_res1 = Anchored_Map(json_file)
Under normal circumstances, the position of the cluster linked to a single anchor point will be outside the circle，if the parameter inner=True, free node position are restricted within the circle. more over, this parameter can be used with center and r，center and r can determine a circle. In this way you can control the position of the node more accurately.
pos_res2 = Anchored_Map(json_file,inner=True) 
the parameter anchor_nodes is passed in an array, you can specify those nodes as anchor points
anchor_nodes = [0,1,2,3,4,5,6,7,8,9,10]
pos_res1 = Anchored_Map(json_file,anchor_nodes=anchor_nodes)