SKSurrogate is a suite of tools focused on designing and tuning machine learning pipelines and
track the evolution of simple modeling tasks.
SKSurrogate is designed to employ Surrogate
Optimization technique in a flexible way which enjoys from the wealth of existing well-known python
tools. It provides a ui to perform surrogate optimization on general functions and has a familiar
ui to perform hyperparameter tuning for scikit-learn compatible models.
SKSurrogate uses surrogate optimization and evolutionary optimization algorithms to
construct and tune complex pipelines (automl) based on a set of given (scikit-learn compatible) models, like
Note that automl part is not designed to result in outstanding results in a few minutes, rather it needs hours if not days to come up with a good result. There are various methods that speeds up the pipeline design/optimization process such as selecting faster surrogates, fewer number of steps and evolutionary optimization settings.
The evolutionary optimization module is designed to be very flexible and can be modified to perform evolutionary optimization on any given evolutionary compatible problem.
To install SKSurrogate, run the following in terminal:
sudo python setup.py install
The documentation is produced by Sphinx and is intended to cover code usage as well as a bit of theory to explain each method briefly. For more details refer to the documentation at sksurrogate.rtfd.io.
This code is distributed under MIT license:
Copyright (c) 2020 Mehdi Ghasemi
Permission is hereby granted, free of charge, to any person obtaining a copy of this software and associated documentation files (the “Software”), to deal in the Software without restriction, including without limitation the rights to use, copy, modify, merge, publish, distribute, sublicense, and/or sell copies of the Software, and to permit persons to whom the Software is furnished to do so, subject to the following conditions:
The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.
THE SOFTWARE IS PROVIDED “AS IS”, WITHOUT WARRANTY OF ANY KIND, EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.