MLOS
MLOS copied to clipboard
microsoft
MLOS is a project to enable autotuning for systems.
We still don't have it implemented. An environment must check when there is a change in the corresponding covariant group (simply by calling `TunableGroup.is_updated("group_name")`) and reset itself, if applicable. For...
* [x] Pass multi-column DataFrame instead of Sequence to `BaseOptimizer.register()` and other methods that deal with scores * [x] Update mlos_bench `MlosCoreOptimizer` to support the new mlos_core API * [x]...
We already have the `ExperimentData.objectives`, but it might be convenient to get that data on the Trial level, too. (Note that the objectives for a trial may be different from...
Reproduced on Windows only so far. To reproduce, create a conda environment: `conda create -n "p312" python=3.12 ipython ConfigSpace` and then run: ```python import ConfigSpace as CS space = CS.ConfigurationSpace(seed=1234)...
Enables changing the defaults in the SMAC optimizer to enable context and multi-fidelity tuning. This requires changes across the project to support the concept of context, as well as updating...
To support a dashboard GUI interface that allows creating new experiment variations from existing config structures ( #732 ), it would be useful for Environment configs to be able to...
- (Storage) APIs to - [ ] create new Experiments - how to get/update configs? git URI? - [ ] set Experiment to runnable - daemon to start mlos_bench process...
Useful for creating graphs in mlos_viz