pymc4
pymc4 copied to clipboard
Experimental PyMC interface for TensorFlow Probability. Official work on this project has been discontinued.
The pymc4.flow subpackage is the core of pymc4, and it is almost completely undocumented! We have to fix this to help potential contributors and users. Furthermore, we need to fix...
The pymc4/distributions/distribution.py module is mostly undocumented. We have to fix this in order to solve #142.
The pymc4/inference/sampling.py module is mostly undocumented. We have to fix this in order to solve #142.
Numpy convention requires that all modules in the pymc4 package have a docstring, potentially a single line at the top of the module. This must be fixed in order to...
It would be very nice to implement a pattern where we can have factories for distributions that can be re-parametrized in terms of scale/loc. A good example would be the...
As pymc4 grows, the test suite gets bigger and we add more fixtures to the test scripts. I think that it would be good practice to start to move the...
I was looking to contribute to PyMC4 and based on #44 I thought I would add the discrete Truncated Normal distribution for a start. When I looked at the source...
I installed tensorflow 2.1 and pymc4 on window 10 machine. The packages are install in a new venv environment. I can import tensorflow (with some warning), but when trying to...
Filing this as an issue so we don't forget: https://github.com/pymc-devs/pymc4/pull/41#issuecomment-453775302 I think we can achieve most of the important distributions using `tfp.distributions.Mixture` and `tfp.distributions.Bijector` (e.g. the zero-inflated distributions are just...
#190 gave us a “basic usage” notebook. In there, @lucianopaz let me know that #180 will let us back-transform values (to non-transformed space) post-sampling. This issue is to serve as...