MLOS Package APIs Overview¶
This is a list of major functions and classes provided by the MLOS packages.
mlos-core API¶
This is a list of major functions and classes provided by mlos_core.
Optimizers¶
|
Enumerate supported MlosCore optimizers. |
Simple factory class for creating BaseOptimizer-derived objects. |
|
|
Create a new optimizer instance, given the parameter space, optimizer type, and potential optimizer options. |
|
Optimizer abstract base class defining the basic interface. |
|
Optimizer class that produces random suggestions. |
|
Wrapper class for FLAML Optimizer: A fast library for AutoML and tuning. |
|
Abstract base class defining the interface for Bayesian optimization. |
|
Wrapper class for SMAC based Bayesian optimization. |
Spaces¶
Converters¶
|
Converts a ConfigSpace.ConfigurationSpace to dict. |
Space Adapters¶
|
Enumerate supported MlosCore space adapters. |
Simple factory class for creating BaseSpaceAdapter-derived objects. |
|
|
Create a new space adapter instance, given the parameter space and potential space adapter options. |
|
SpaceAdapter abstract class defining the basic interface. |
|
Identity (no-op) SpaceAdapter class. |
|
Implementation of LlamaTune, a set of parameter space transformation techniques, aimed at improving the sample-efficiency of the underlying optimizer. |
mlos-bench API¶
This is a list of major functions and classes provided by mlos_bench.
Main¶
The script to run the benchmarks or the optimization loop.
Also available as mlos_bench command line tool.
Note
The are json config examples and json schemas on the main source code repository site.
Benchmark Environments¶
|
Enum for the status of the benchmark/environment. |
|
An abstract base of all benchmark environments. |
|
Composite benchmark environment. |
|
Scheduler-side environment to mock the benchmark results. |
Local Environments¶
|
Scheduler-side Environment that runs scripts locally. |
|
Scheduler-side Environment that runs scripts locally and uploads/downloads data to the shared file storage. |
Remote Environments¶
|
Environment to run benchmarks and scripts on a remote host OS. |
|
OS Level Environment for a host. |
|
Remote VM/host environment. |
|
Remote host environment. |
Tunable Parameters¶
|
A tunable parameter definition and its current value. |
|
A collection of covariant groups of tunable parameters. |
Service Mix-ins¶
|
An abstract base of all Environment Services and used to build up mix-ins. |
|
An abstract base of all file shares. |
|
Collection of methods to deserialize the Environment, Service, and TunableGroups objects. |
Local Services¶
|
Collection of methods to run scripts and commands in an external process on the node acting as the scheduler. |
Remote Azure Services¶
|
Helper methods to manage VMs on Azure. |
|
Helper methods for interacting with Azure File Share. |
Optimizer Adapters¶
|
An abstract interface between the benchmarking framework and mlos_core optimizers. |
|
Mock optimizer to test the Environment API. |
|
A wrapper class for the mlos_core optimizers. |
Storage¶
Base Runtime Backends¶
|
An abstract interface between the benchmarking framework and storage systems (e.g., SQLite or MLFLow). |
|
Create a new storage object from JSON5 config file. |
SQL DB Storage Backend¶
|
An implementation of the Storage interface using SQLAlchemy backend. |
Analysis Client Access APIs¶
|
Base interface for accessing the stored experiment benchmark data. |
|
Base interface for accessing the stored experiment benchmark trial data. |
|
Base interface for accessing the stored experiment benchmark (tunable) config data. |
|
Base interface for accessing the stored experiment benchmark tunable config trial group data. |
mlos-viz API¶
This is a list of major functions and classes provided by mlos_viz.
|
What method to use for visualizing the experiment results. |
|
Plots the results of the experiment. |