As long as Theano itself still works in versions of Python that SymPy supports, it would be better to deprecate the function rather than just remove it. The function theano_code is public API so just removing it would be a backwards compatibility break, which we try to avoid without deprecations. See https://github.com/sympy/sympy/wiki/Deprecating-policy
As to whether it should be deprecated or just left alone, I guess it depends. To me it would make sense to deprecate it if theano truly is a dead project. Is Aesara a drop-in replacement for Theano? Why was a fork done, instead of just continuing development on Theano itself? The symbolic work on Aesara looks interesting. To be clear, I am not at all opposed to supporting Aesara in SymPy. My only concern is with the backwards compatibility break from removing support for Theano. As an aside, one thing that has always annoyed me with Theano, especially when SymPy imports it for its tests, is that Theano does a lot of things at import time, such as running compilers to see if they work and printing many warnings (see https://github.com/sympy/sympy/pull/14747). I think it would be better to do these things at runtime, e.g., after the first function is called. I hope Aesara can improve this situation over Theano. Aaron Meurer On Mon, Mar 15, 2021 at 3:42 PM Brandon T. Willard <[email protected]> wrote: > > I'm the lead developer/manager of the Aesara project and a core developer in > the PyMC group. > > Theano has officially stopped development and currently exists in an > effectively frozen state. As stated in Theano's README, its successor is our > fork, Aesara. > > If there are users who still rely on legacy Theano and SymPy's Theano > features, then it would make sense to maintain the existing Theano code. I'm > not personally aware of any projects that do, so I can't say, but I do know > that Aesara has considerably more potential for developing SymPy interactions > (e.g. see some preliminary symbolic work here) in the nearterm. > On Monday, March 15, 2021 at 3:53:02 PM UTC-5 [email protected] wrote: >> >> I would also like more context on what the status of Theano and Aesara >> is. The pull request as it is breaks compatibility by renaming the >> function and making it no longer work with theano. I think it would be >> better to add a new function, aesara_code, and keep the old >> theano_code() intact. I'm not clear if it should also be deprecated. >> Theano is currently installed in the tests, meaning it presumably >> still works. Do we know at what point it will stop working? >> >> Perhaps someone can forward this thread to one of the Aesara >> developers so they can comment here. >> >> Aaron Meurer >> >> On Mon, Mar 15, 2021 at 2:16 PM Oscar Benjamin >> <[email protected]> wrote: >> > >> > Hi all, >> > >> > Does anyone use SymPy with Theano? >> > >> > There is a PR which proposes to drop Theano support in favour of Aesara: >> > https://github.com/sympy/sympy/pull/21087 >> > >> > The PR looks fine to me but I don't actually know what Theano/Aesara >> > are. Apparently Theano is no longer actively maintained and Aesara is >> > a fork that should be considered a natural successor. >> > >> > If anyone has any thoughts on this then please say so (here or on the >> > PR). Otherwise I'm inclined to merge the PR because the author clearly >> > knows more than I do about this and says that this is a reasonable >> > approach. >> > >> > Oscar >> > >> > -- >> > You received this message because you are subscribed to the Google Groups >> > "sympy" group. >> > To unsubscribe from this group and stop receiving emails from it, send an >> > email to [email protected]. >> > To view this discussion on the web visit >> > https://groups.google.com/d/msgid/sympy/CAHVvXxTCV3U2YxE5%3D3-us5RbtkfF8ZaqmvKA-8oM9QkPdXJUxg%40mail.gmail.com. > > -- > You received this message because you are subscribed to the Google Groups > "sympy" group. > To unsubscribe from this group and stop receiving emails from it, send an > email to [email protected]. > To view this discussion on the web visit > https://groups.google.com/d/msgid/sympy/d079659c-e249-4d22-a231-8caa2a101716n%40googlegroups.com. -- You received this message because you are subscribed to the Google Groups "sympy" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/sympy/CAKgW%3D6JcBTU1phLgs0frjt7uOkErc7xEFFXbbizzCZyGLgq0oA%40mail.gmail.com.
