- First release on PyPI.
- NEW: Includes primary optimization techniques such as global-best PSO and local-best PSO - #1, #3
- FIX: Patch on LocalBestPSO implementation. It seems that it's not returning the best value of the neighbors, this fixes the problem .
- NEW: Test functions for single-objective problems - #6, #10, #14. Contributed by @Carl-K. Thank you!
- NEW: Binary Particle Swarm Optimization - #7, #17
- FIX: Fix on Ackley function return error - #22
- IMPROVED: Documentation and unit tests - #16
- FIX: Added a patch to fix
pip
installation
- NEW: easy graphics environment. This new plotting environment makes it easier to plot the costs and swarm movement in 2-d or 3-d planes - #30, #31
- NEW: Native GridSearch and RandomSearch implementations for finding the best hyperparameters in controlling swarm behaviour - #4, #20, #25. Contributed by @SioKCronin. Thanks a lot!
- NEW: Added tests for hyperparameter search techniques - #27, #28, #40. Contributed by @jazcap53. Thank you so much!
- IMPROVED: Updated structure of Base classes for higher extensibility
- FIX: Fixed patch on
local_best.py
andbinary.py
- #33, #34. Thanks for the awesome fix, @CPapadim! - NEW: Git now ignores IPython notebook checkpoints
- NEW: PySwarms is now published on the Journal of Open Source Software (JOSS)! You can check the review here. In addition, you can also find our paper in this link. Thanks a lot to @kyleniemeyer and @stsievert for the thoughtful reviews and comments.
- NEW: You can now set the initial position wherever you want - #93
- FIX: Quick-fix for the Rosenbrock function - #98
- NEW: Tolerance can now be set to break during iteration - #100
Thanks for all the wonderful Pull Requests, @mamadyonline!
- NEW: New PySwarms backend. You can now build native swarm implementations using this module! - #115, #116, #117
- DEPRECATED: Drop Python 2.7 version support. This package now supports Python 3.4 and up - #113
- IMPROVED: All tests were ported into pytest - #114
- FIX: Fix sigmoid function in BinaryPSO - #145. Thanks a lot @ThomasCES!
- NEW: New topologies: Pyramid, Random, and Von Neumann. More ways for your particles to interact! - #176, #177, #155, #142. Thanks a lot @whzup!
- NEW: New GeneralOptimizer algorithm that allows you to switch-out topologies for your optimization needs - #151. Thanks a lot @whzup!
- NEW: All topologies now have a static attribute. Neigbors can now be set initially or computed dynamically - #164. Thanks a lot @whzup!
- NEW: New single-objective functions - #168. Awesome work, @jayspeidell!
- NEW: New tutorial on Inverse Kinematics using Particle Swarm Optimization - #141. Thanks a lot @whzup!
- NEW: New plotters module for visualization. The environment module is now deprecated - #135
- IMPROVED: Keyword arguments can now be passed in the
optimize()
method for your custom objective functions - #144. Great job, @bradahoward
- NEW: New collaboration tool using Vagrantfiles - #193. Thanks a lot @jdbohrman!
- NEW: Add configuration file for pyup.io - #210
- FIX: Fix incomplete documentation in ReadTheDocs - #208
- IMPROVED: Update dependencies via pyup - #204
- NEW: The console output is now generated by the
Reporter
module - #227 - NEW: A
@cost
decorator which automatically scales to the whole swarm - #226 - FIX: A bug in the topologies where the best position in some topologies was not calculated using the nearest neighbours - #253
- FIX: Swarm init positions - #249 Thanks @dfhljf!
- IMPROVED: Better naming for the benchmark functions - #222 Thanks @nik1082!
- IMPROVED: Error handling in the
Optimizers
- #232 - IMPROVED: New management method for dependencies - #263
- DEPRECATED: The environments module is now deprecated - #217
This is the first major release of PySwarms. Starting today, we will be adhering to a better semantic versioning guidelines. We will be updating the project wikis shortly after. The maintainers believe that PySwarms is mature enough to merit a version 1, this would also help us release more often (mostly minor releases) and create patch releases as soon as possible.
Also, we will be maintaining a quarterly release cycle, where the next minor release (v.1.0.0) will be on June. All enhancements and new features will be staged on the development
branch, then will be merged back to the master
branch at the end of the cycle. However, bug fixes and documentation errors will merit a patch release, and will be merged to master
immediately.
- NEW: Boundary and velocity handlers to resolve stuck particles - #238 . All thanks for our maintainer, @whzup !
- FIX: Duplication function calls during optimization, hopefully your long-running objective functions won't take doubly long. - #266. Thank you @danielcorreia96 !
- FIX: Handlers memory management so that it works all the time - #286 . Thanks for this @whzup !
- FIX: Re-introduce fix for multiple optimization function calls - #290 . Thank you once more @danielcorreia96 !
- FIX: BinaryPSO should return final best position instead of final swarm - #293 . Thank you once more @danielcorreia96 !
This new version adds support for parallel particle evaluation, better documentation, multiple fixes, and updated build dependencies.
- NEW: Updated API documentation - #344
- NEW: Relaxed dependencies when installing pyswarms - #345
- NEW: We're now using Azure Pipelines for our builds! - #327
- NEW: Add notebook for electric circuits - #288 . Thank you @miguelcocruz !
- NEW: Parallel particle evaluation - #312 . Thahnk you once more @danielcorreia96 !
- FIX: Fix optimise methods returning incorrect best_pos - #322 . Thank you @ichbinjakes !
- FIX: Fix SearchBase parameter - #328 . Thank you @Kutim !
- FIX: Fix basic optimization example - #329 . Thank you @IanBoyanZhang !
- FIX: Fix global best velocity equation - #330 . Thank you @craymichael !
- FIX: Update sample code to new API - #296 . Thank you @ndngo !