Follow up question regarding github Ray Issue 6144

Ray Libraries (Data, Train, Tune, Serve)
1

OS Platform and Distribution: Kubuntu 18.04
Ray installed from (source or binary): Installed binary via pip install
Ray version: 2.5.1
Python version: 3.11.4

Hi,

I have been using ray tune 0.7.7 with Python 3.6.14 for a long time in the way as described in [tune] Unexpected num_samples and trial execution (setting of self.config) behavior #6144, but need to upgrade to the latest ray tune (2.5.1) - other updated environment items would be as noted at the top of this particular comment. @richardliaw (or anyone else), could anyone please provide me with any pointers regarding how to achieve the same behaviour as described above, but for the latest ray (2.5.1), because I can see ray tune has changed in a number of ways since 0.7.7, Many thanks for any pointers anyone can provide.

Thanks
Andrew|

2

Hi @andrewv99, most of the behavior should be the same as in the old Ray Tune version. Perhaps you can try to run it, and if you run into any problems, post your code and the experienced problems here?