-
Notifications
You must be signed in to change notification settings - Fork 1.7k
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
Upgrade torch to 1.13.0 #3683
Comments
by Mayank: In case someone is interested in reading further about torch-2.0 and torch-1.13, |
huggingface/transformers#20168 This is how hf-transformers solved their issue. Seems a bit untidy solution for us. But still investigating. |
Just to throw my two cents here. We would like to not to temporarily drop support for TAPAS models since we are actively developing a German version to use for clients. The TAPAS models are used for the TableReader. |
@sjrl pointed me to this solution for getting rid of torch-scatter dependency. huggingface/transformers#20101 @julian-risch this is on the similar line we discussed! Hopefully next release we will be able to upgrade the torch to 1.13 |
I've opened this PR #3703 bumping the transformers version which allows us to remove torch-scatter as a dependency for Haystack. This should hopefully help us be able to bump to Pytorch 1.13! |
**Describe the issue
Let's upgrade to torch 1.13.0 (and to 1.13.1 as soon as it is released) to prepare for speed improvements coming with version 2.0.
Colab notebooks also have version 1.13.0 pre-installed now.
@mayankjobanputra tracked changes between 1.13.1 and 2.0. It seems that the changes from there aren't big. If we upgrade to 1.13 now it will hopefully allow us to make the jump to 2.0 quickly.
**Additional context
Add any other context about the problem here.
The text was updated successfully, but these errors were encountered: