Skip to content

Missing conv documentation #275

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

Open
thipokKub opened this issue Apr 13, 2023 · 4 comments · May be fixed by #548
Open

Missing conv documentation #275

thipokKub opened this issue Apr 13, 2023 · 4 comments · May be fixed by #548
Labels

Comments

@thipokKub
Copy link

Issue with current documentation:

I was looking for convolutional operator, but the PyTensor's document seems to be missing

Idea or request for content:

Add how to use conv operator in the document

@thipokKub thipokKub added the docs label Apr 13, 2023
@thipokKub
Copy link
Author

From this question. I use the pytensor.tensor.conv module and got the following error

NotImplementedError: No JAX conversion for the given `Op`: AbstractConv2d{convdim=2, border_mode=(0, (2, 1)), subsample=(1, 1), filter_flip=True, imshp=(None, None, None, None), kshp=(None, None, None, None), filter_dilation=(1, 1), num_groups=1, unshared=False}

It seems to related to another question. Jax support convolution, but the conversion from pytensor seems to be lacking. I am not sure if it is a direct mapping or not?

@jessegrabowski
Copy link
Member

The original Aesara (Theano?) API docs for the tensor.nnet.conv sub-module are on the wayback machine here:

https://web.archive.org/web/20220519051518/https://aesara.readthedocs.io/en/latest/library/tensor/nnet/conv.html

Could do a copy/paste job? I know even less about docs than anything else.

@ricardoV94
Copy link
Member

They are certainly still there in the git history ;)

@jessegrabowski
Copy link
Member

I thought I was so clever : (

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
Projects
None yet
Development

Successfully merging a pull request may close this issue.

3 participants