Pyspi

Introduction
This patch release brings a few minor updates including a new high contrast logo for dark mode users, improved SPI unit testing (with a new benchmarking dataset) and fixes for potential security vulnerability issues.

Highlights of this release
- New high contrast logo for dark-mode users.
- Improved SPI unit testing with z-scoring approach to flag SPIs with differing outputs.
- New coupled map lattice (CML) benchmarking dataset.
- Fix for potential security vulnerability issues in scikit-learn.

What's Changed
- Replaced the old `standard_normal.npy` benchmarking dataset with a coupled map lattice (`cml7.npy`), along with its associated .pkl file containing the benchmark values (`CML7_benchmark_tables.pkl`) generated in a fresh Ubuntu environment.
- Updated the README to automatically select either the regular or new dark mode logo based on the user's theme.
- Added new `conftest.py` file for pytest to customise the unit testing outputs.
- Added a new `pyproject.toml` file for configuring the package for publishing to PyPI.

New features
- Improved SPI unit testing with a new coupled map lattice benchmarking dataset (`cml7.npy`) consisting of 7 processes and 100 observations per process.
- Z-scoring approach in unit testing pipeline to flag potential changes in SPI outputs as a result of algorithmic changes, etc. SPIs with outputs differing by more than a specified threshold are "flagged" and summarised in a table.
- Added a darkmode _pyspi_ logo to the README which is shown for users with the dark-mode GitHub theme.

Bug Fixes
- Fixed a scikit-learn security vulnerability issue with severity "high" (pertaining to denial of service) by upgrading scikit-learn from version `0.24.1` to version `1.0.1`.
- Fixed Int64 deprecation issue (cannot import name `Int64Index` from `pandas`) by fixing pandas to version `1.5.0`.
- Fixed unknown character issue for Windows users resulting from not specifying an encoding when loading the "README" in `setup.py`. Now fixed to `utf-8` for consistency across platforms.

Introduction
PySPI v0.4.1 introduces several minor changes to the existing README, as well as migrating documentation from "readthedocs" to an all new "GitBook" page. Simple unit testing has also been incorporated for each of the SPIs using a benchmarking dataset to check for the consistency of outputs.

Highlights of this release

What's Changed
- Removal of old /docs directory
- Addition of a /tests directory for unit testing
- Updated README
- Addition of CODE_OF_CONDUCT.md and SECURITY.md

New features
- Basic unit testing incorporated into a GitHub Actions workflow.
- Updated README file with links to the new GitBooks hosted documentation to replace the old "readthedocs" documentation.
- Added a code of conduct markdown
- Added a security policy markdown

Bug Fixes
- Fixed a PyTorch security vulnerability issue with severity "critical" (pertaining to arbitrary code execution) by updating torch from version `1.10.0` to `1.13.1`.

- The directed info measure now uses entropy rate in its calculation to closer resemble the streaming method described in literature.
- The code (mostly) uses black formatting now for readability

pynats-v0.1
This release is the version that was used for computing the results in the paper.

- Included fast compute option for the calculator (~10x speedup for bivariate time series)
- Minor bug fixes

Code works on demo and has all of the statistics referenced in our paper, plus another ~100 more.