MLOS
MLOS copied to clipboard
microsoft
MLOS is a project to enable autotuning for systems.
Summary of changes: * Pass optional weights for optimization targets in mlos_core * Implement weighted average for multi-objective optimization in FLAML * Add more unit tests for multi-objective optimization on...
There may still be some deployment use cases for this so probably not just yet. When we do get to it, don't forget to look for various `sys.version_info` checks to...
Previously this project used autopep8. This did not automatically enforce many aspects of style causing many files to use different styles. Additionally not all standards in pep8 were being enforced....
Some non-target benchmark metrics can be useful for optimization. The user should be able to specify which context metrics to pass to the optimizer.
Related to #403 we should also capture additional metadata about VMs automatically where ever possible.
`is_defaults` looks like a boolean that is stored in the results storage when a new trial is created. However, the current implementation is storing a serialized method that tends exceeds...
This is part of PR #777. Merge after ~#779~ and ... (azcopy PR coming soon)
Make `AzureFlieShareService` class use token-based credential instead of the access key. This PR is part of #777 **NOTE:** * We have to use late initialization of the `_share_client` to avoid...