I need to plot the graphs using networkx . How can we do that without using groupby as it converts the data to panda series

Question

I have data in a CSV file. Which contains data of all students and registered courses. After reading the data I selected 2 columns i.e ID and course_title. I need to plot a graph for each class where nodes are student IDs and each node should be connected to every other node in the class. So I need to pair students in a class for that I have written below code but it didn't work. Can you please correct me where am I doing wrong?

df_node = cs_class_df[['ID','COURSE']]

df_node = df_node.groupby(["COURSE"])

group_classes = df_node["ID"].apply(list)

group_classes = group_classes.reset_index()

print(group_classes)

group_classes= group_classes.loc[4:4]

print(group_classes)

df_list = group_classes["ID"]

df_list actually contains [1,4,5,8] df_list has become a series so even If I convert this to array its length is 1. I need to pair the elements in df_list like [(1,4),(1,5),(1,8),(4,5),(4,8),(5,8)]