Update pandas_df_utils.py

[yueguoguo]

Fixed the issues of the negative sampler

Description

Fix the issue in the negative sampling function

Related Issues

N/A

Checklist:

• I have followed the contribution guidelines and code style for this project.
• I have added tests covering my contributions.
• I have updated the documentation accordingly.
• This PR is being made to staging branch and not to main branch.

[anargyri]

I think we can make negative sampling more efficient. Currently we create all user item pairs, which is time and memory consuming i.e. O(n_users x n_items). Instead, we could sample indexes in O(n_sample) time e.g. using numpy sampling functions.

[miguelgfierro]

I think we can make negative sampling more efficient. Currently we create all user item pairs, which is time and memory consuming i.e. O(n_users x n_items). Instead, we could sample indexes in O(n_sample) time e.g. using numpy sampling functions.

+ 1 to using numpy