We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
In terms of explaining RecoGym http://localhost:8888/notebooks/Bandit%20Feedback%20-%20Evaluate%20Organic%20Model%20with%20Bandit%20Signal.ipynb# is one of the best notebooks but for some reason the verify_IPS seems to be returning nan on the random agent.
It typically produces a value of 0.
This bug seems to have been introduced when the option of memory saving by not logging ps-a was introduced and possibly line 776 in : https://github.com/criteo-research/reco-gym/blob/master/recogym/evaluate_agent.py
The good news is that ps-a is used only by these evaluate calls which are great in this notebook but not usually needed.
The text was updated successfully, but these errors were encountered:
No branches or pull requests
In terms of explaining RecoGym http://localhost:8888/notebooks/Bandit%20Feedback%20-%20Evaluate%20Organic%20Model%20with%20Bandit%20Signal.ipynb#
is one of the best notebooks but for some reason the verify_IPS seems to be returning nan on the random agent.
It typically produces a value of 0.
This bug seems to have been introduced when the option of memory saving by not logging ps-a was introduced and possibly line 776 in : https://github.com/criteo-research/reco-gym/blob/master/recogym/evaluate_agent.py
The good news is that ps-a is used only by these evaluate calls which are great in this notebook but not usually needed.
The text was updated successfully, but these errors were encountered: