First non-alpha release! Featuring:

• No memory leaks! 😵
• Beam search algorithm 🥳
• New building setup

And (probably not) much more! 💥

• Added the beam_search algorithm.
• Extended Protein and AminoAcid with copy, assign, and print overloaded functions.
• Added a new dataset where each CSV contains proteins with a fixed interval of H-ratio.
• Reworked the setuptools scripts to build from pyproject.toml.
• Added different prune criteria to depth_first_search_bnb.

Note that the documentation has not been updated.

Cleaned up the core code for generating proteins.
Fixed creating visualizations, as it previously was based on the old core code.
Added a "paper" style for visualizations, which only generates a conformation without any axis or labels.

This release contains small hotfixes for the v0.2-alpha.3 release.

• Fixed versioning in the GitHub actions pipeline.
• Added AminoAcid, load_vanEck250, and load_vanEck1000 to the import path of prospr.
• Added tests for the datasets.
• Removed support for Python 11 as it requires a different setup of the pyproject.toml file.

Many new functionalities, which are not documented correctly yet.

• The Protein objects now consist of AminoAcid objects.
• The AminoAcid objects form the internal linkage, while the Protein objects manage the total space for folding.
• The internal workings of Protein have been fully re-designed and function calls are now easier.
• A Pytest test suite has been made for the workings of Protein and the algorithms.
• There is a core test suite written in C++ to test the core without building the Python module.
• Upped the minimum required Python version to 3.9.
• GitHub actions versions have been increased, and we now build for specific Python versions.
• Documentation was added, but requires updating.
• Prospr now has its own logo!

Prospr is one step closer to reaching beta! New functionality:

• Added datasets to use
• Added archives of the C++ core and datasets
• Optimized CI-CD pipeline
• Optimized develop environment with pre-commit and hooks for automatically generating the archives

Prospr can now actually be used as intended! Core functionality can be directly imported from main package and visualization functions are under prospr.visualize. Next steps will be streamlining everything, adding tests, and adding documentation.

Changes:

• Fixed import structure
• Fixed incompatibility between the new core bindings and the visualization & helpers code

Added Github Actions driven CI/CD and pip support for Ubuntu, macOS, and Windows-64bit.