PyBayes Change Log

This file mentions changes between PyBayes versions that are important for its users. Most recent versions are mentioned on top.

Changes between 0.3 and 0.4

• Filter.posterior‘s signature was changed to return CPdf instead of Pdf. It is more versatile - it for example allows you you to return zero-condition ProdCPdf.
• The optimised Cython PyBayes version was turned to use Cython memoryviews (with help from Ceygen) instead of NumPy arrays where it makes sense. Cython memoryviews and NumPy arrays are mostly mutually compatible, but you may need to use np.asarray(...) or convert your code to use e.g. np.sum(pybayes_result) instead of pybayes_result.sum(). Alternatively, you can convert your code to use Cython memoryviews too.
• Use of bundled Tokyo is replaced by the Ceygen project and Tokyo submodule is removed.
• ParticleFilter.bayes() now ignores cond completely. Yell if you need it.
• ParticleFilter lost last emp_pdf argument. Pass the same object as the init_pdf argument to achieve the same thing.
• Test- and stress-suite no longer need PyBayes to be installed. (no privilege problems etc.)
• Build-system was rewritten so that it is no longer an ugly hack. .pxd and .py files are now installed along .so (.dll) files for interoperability and additional openness. Better parsing of setup.py arguments and custom parameters visible in the –help command.
• (C)Pdf shape() and cond_shape() functions are no longer abstract and just return self.rv.dimension and self.cond_rv.dimension respectively. CPdf subclasses therefore should not implement these methods. This is a backwards compatible change API-wise.