Xavier Bouthillier
Xavier Bouthillier
Make sure to include tests on proper error messages if fidelity dimension is not included.
I agree that would be much simpler. The `Params` object was a sql-like way of representing the data in the database which never proved useful and will probably never will....
Must adapt doc after merging #957
@notoraptor Do you remember if Oríon was installed in editable mode?
https://github.com/Epistimio/orion/blob/develop/src/orion/testing/algo.py#L405 https://github.com/Epistimio/orion/blob/develop/src/orion/testing/algo.py#L421
Yeah our position was that we were making an MVP and max/min was quite easy for the users to handle so no need to include in Oríon. I think Oríon...
For reference, commandline call is ```bash orion hunt -c orion_bayesian_config.yaml \ python 3 test_orion_2.py Breakout 4 "/some/path" "/some/other/path.yml" \ --learning_rate~'loguniform(1e-6, 1e-3)' --alpha~'loguniform(0.94, 0.96)' ``` Next time if you can copy...
Yes, config here would be to specify another storage than the one in global configuration, if necessary.
If there are experiment ID conflicts during `load`, we could offer to `ignore`, `overwrite` or `bump version of incoming experiment`. When overwritting, should we deleting the prior trials? :thinking:
Thanks a lot for documentation the flow of execution during a call to `orion hunt`! I think it is quite complete, despite maybe a short mention of experiment version check/experiment...