Brandon T. Willard

Results 358 comments of Brandon T. Willard

I've removed the bug label, because I can't find an actual error caused by this behavior. It's really just "unintuitive" at the user-level and will likely trip people up.

The most basic algebraic operators (e.g. `__[r]add__`, `__[r]sub__`, etc.) have been added to the TensorFlow meta tensor class in #79. The same needs to be done for Theano, along with...

There are some ambiguities regarding when/how a term should be constructed by the aforementioned constraint, and that leads me to believe that an entirely different goal should be designated to...

> Would using Hy have any advantage if we can translate between code and sexprs? Hy would be convenient for a number of reasons (e.g. manipulating AST), but a lot...

This addition should follow the basic design/approach/template of `pymc3_hmm.step_methods.FFBSStep` (e.g. implement all the same methods, perform similar (symbolic) input validation and setup, have an informative docstring with some math details,...

Looks like someone already has [a PyMC3 implementation](https://gist.github.com/dadaromeo/33e581d9e3bcbad83531b4a91a87509f)!

#81 added some important improvements, including two forms of in-place updating for series-length arrays; however, there are still a couple important in-place opportunities that involved the Theano graphs (see the...

Can you provide a complete example? That will help me understand the context better.

I think I understand, and it sounds good, but I'll need a solid minute to think about the module in which this helper function should go, and so forth. Otherwise,...

Sorry, I've been extremely busy. Anyway, these look good, so I gladly invite you to create a PR for them.