aws / aws/amazon-sagemaker-examples
Reward and Regret calculation duplicated in bandits_recsys_movielens_testbed example ?
Open
- Dominant language
- Jupyter Notebook
- Stars
- 11k
- Forks
- 7k
- Avg merge
- 8h 29m
- Merged PRs (30d)
- 8
Description
Hi, is the rewards and regret calculation necessary two times ?
1) In the step function of src/env.py, line number 97-99.
2) In train.py, line 101.
In both places, rewards and regrets are being calculated based on same set of actions.
Contributor guide
Assessment
This issue has not been assessed yet.