David Pitt
David Pitt
Converting to a discussion since no PR has been opened. Please feel free to suggest a fix!
Training logging is controlled by the parameter `verbose`, which is set to False by default in the Trainer module. It can be turned on to log training
Fixed this in #365. Pull the dev version of the library to check it out! New pip release soon too
Are there plans to introduce a full model using the CodaBlocks, or to add a training recipe for CODANO? On Wed, Oct 2, 2024 at 11:30 AM Md Ashiqur rahman...
Superseded by #462 (merged)
This is great, thanks so much @zongyi-li ! Comments to follow but this looks really good. One note: the online runner does not haver access to the 128x128 darcy-flow dataset...
> My main question is whether there shouldn't be some more separation between the concepts (the algorithm) and the problem specific implementation (for Darcy, Burgers, etc). Perhaps this is just...
Fixed the issues in the doc and in `neuralop.models.UQNO` causing tests to fail
No, this was pending some kind of fix to the sub module saving procedure On Wed, Jul 2, 2025 at 4:59 AM Jean Kossaifi ***@***.***> wrote: > *JeanKossaifi* left a...
Yes, the UQNO wouldn't properly serialize. I sent a stub to reproduce the issue and we were deciding whether to change the basemodel interface On Tue, Sep 9, 2025 at...