Hello,
For sample efficiency approach I need to randomly reevaluate target policy/value and update it.
The client API allow to remove such datas but not the tf_client. As the core functions allow it I suggest to bind parameters.
Thanks,
Thanks a lot for contributing to Reverb and sorry for the long delay. If you could take a look at the merge conflicts that have developed then I'd be more than happy to accept this PR!
Hello, For sample efficiency approach I need to randomly reevaluate target policy/value and update it. The client API allow to remove such datas but not the tf_client. As the core functions allow it I suggest to bind parameters. Thanks,