-
Notifications
You must be signed in to change notification settings - Fork 134
Extend inverse @ b as solve rewrite to batched dots #945
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Comments
@ricardoV94 I am currently working on this issue and I have some doubts. As I see it the |
@AdvH039 you're right, the concern is that the rewrite currently doesn't apply with batch Dot. You don't have to worry about Batch Dot22, as Dot22 is a specialized op that shouldn't ever be batched (it is only introduced when there's no batching going on). |
Hi @ricardoV94, |
I've missed it, it's better to keep the development discussion here on github. |
@ricardoV94 Ok thanks. Please let me know if there are any provisions for 175hr and what ideas are you looking for as I want to start preparing the proposal as soon as possible. |
@ricardoV94 I have started the discussion and introduced myself. Thanks!! |
I meant the discussion for the issue should happen here, not for GSOC projects |
Oh I'm sorry about that should I take it off? Thanks.(Also are there any developments regarding the gsoc project?) |
This rewrite only triggers for non-blockwise dots, but there's no reason not to also cover Blockwise dots with a Blockise Solve (if solve is not blockwised by default already)
pytensor/pytensor/tensor/rewriting/linalg.py
Line 90 in ad27dc7
The text was updated successfully, but these errors were encountered: