Kto najlepiej ćwierka? Podstawy Graphframes + Tweepy

import tweepy
import csv

class FollowersFinder:
    def __init__(self, consumer_key, consumer_secret, access_token, access_secret):
        self.scanned_accounts = set()
        auth = tweepy.OAuthHandler(consumer_key, consumer_secret)
        auth.set_access_token(access_token, access_secret)
        self.api = tweepy.API(auth, wait_on_rate_limit=True, wait_on_rate_limit_notify=True)

    def start(self, depth = 3):
        self.max_depth = depth
        account_self_id = self.api.me().id
        self.append_account(account_self_id)
        self.find_relationships(account_self_id)

    def find_relationships(self, account_id, depth = 0):
        if (account_id in self.scanned_accounts):
            return
        if (depth > self.max_depth):
            return
        try:
            followers_ids = self.api.followers_ids(account_id)
            friends_ids = self.api.friends_ids(account_id)
        except tweepy.TweepError:
            print("Failed to run the command on that user, Skipping...")
            return
        self.scanned_accounts.add(account_id)
        if (len(followers_ids) > 10000 or len(friends_ids) > 10000):
            return
        for follower_id in followers_ids:
            self.append_account(follower_id)
            self.append_relationship((follower_id,account_id))
            self.find_relationships(follower_id,depth + 1)
        for friend_id in friends_ids:
            self.append_account(friend_id)
            self.append_relationship((account_id,friend_id))
            self.find_relationships(friend_id,depth + 1)

    def append_account(self, value):
        with open('accounts.csv', mode='a', newline='') as employee_file:
            employee_writer = csv.writer(employee_file, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
            employee_writer.writerow([value])

    def append_relationship(self, value):
        with open('relationships.csv', mode='a', newline='') as employee_file:
            employee_writer = csv.writer(employee_file, delimiter=',', quotechar='"', quoting=csv.QUOTE_MINIMAL)
            employee_writer.writerow(value)


consumer_key = 'srutu'
consumer_secret = 'tutu'
access_token = 'majtki'
access_secret = 'z drutu'

finder = FollowersFinder(consumer_key, consumer_secret, access_token, access_secret)
finder.start(3)

export SPARK_HOME=/home/maciej/spark-2.4.4-bin-hadoop2.7
export HADOOP_HOME=/home/maciej/hadoop-3.2.1
export PATH="$SPARK_HOME/bin:$PATH"
export PYSPARK_DRIVER_PYTHON=jupyter
export PYSPARK_DRIVER_PYTHON_OPTS='notebook'

pyspark --packages graphframes:graphframes:0.7.0-spark2.4-s_2.11

from pyspark.sql import SparkSession
import pyspark.sql.functions as F
import graphframes as G
spark = SparkSession\
        .builder\
        .appName("tweets")\
        .getOrCreate()
accounts = spark.read.csv("accounts_followers.csv")
relationships = spark.read.csv("relationships_followers.csv")
vertices = accounts.select(F.col("_c0").alias("id")).distinct()
edges = relationships\
            .select(
                F.col("_c0").alias("src"),\
                F.col("_c1").alias("dst"),\
                F.lit("follows").alias("relationship"))\
            .distinct()
g = G.GraphFrame(vertices, edges)