Is there an efficient way to calculate the personalized page rank over a huge graph using the networkx library?

Question

I want to calculate the personalized page rank value of every node in a large dense graph, with over a million nodes. Currently I am using the networkx library and setting the personalization value as given in the answer in Using python's networkX to compute personalized page rank.

However, is there a more efficient way, in terms of time complexity, to do this? Thank you very much.

def findRank(graph, node):
    try:
        graph.has_node == False
    except Exception as e:
        print("No such node in the graph")

    nodeList = list(graph.nodes)
    nodeDict = dict()
    for graphnode in nodeList:
        if graphnode != node:
            nodeDict[graphnode] = 0
        else:
            nodeDict[graphnode] = 1
    rank = pagerank_numpy(
        graph,
        alpha=0.85,
        personalization=nodeDict,
        weight='weight',
        dangling=None
    )