Skip to content
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.

Sign up for GitHub

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

Jump to bottom

remove datapoints compatibility for prototype datasets #7154

Merged
merged 4 commits into from
Feb 10, 2023
Merged

remove datapoints compatibility for prototype datasets #7154

merged 4 commits into from
Feb 10, 2023

Conversation

pmeier
Copy link
Collaborator

@pmeier pmeier commented Jan 31, 2023 •
edited by pytorch-bot bot
Loading

This is the same idea as #6663 (comment). Whatever we decide there, can also be applied here.

cc @bjuncek

@pmeier pmeier mentioned this pull request Jan 31, 2023
Closed
@pmeier pmeier marked this pull request as draft February 3, 2023 15:51
@pmeier
Copy link
Collaborator Author

pmeier commented Feb 3, 2023 •
edited
Loading

If we move forward with #7170 (and it looks like it so far), this PR can be reduced to converting all non-datapoint numerical parts of the sample to a plain torch.Tensor and making sure that an image is coming first in the sample.

@pmeier pmeier changed the title move datapoints compat to prototype datasets remove datapoints compatibility for prototype datasets Feb 10, 2023
@pmeier pmeier marked this pull request as ready for review February 10, 2023 14:25
@pmeier
Copy link
Collaborator Author

pmeier commented Feb 10, 2023

I've run the prototype datasets tests locally and they are passing.

@pmeier pmeier requested a review from NicolasHug February 10, 2023 14:26
@pmeier pmeier mentioned this pull request Feb 10, 2023
Closed
49 tasks
NicolasHug
NicolasHug approved these changes Feb 10, 2023
View reviewed changes
Copy link
Member

@NicolasHug NicolasHug left a comment •
edited
Loading

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Stamping - happy with the changes made to the DataPoint class, I haven't looked at the changes made to the prototype datasets though (not critical for now)

@pmeier pmeier merged commit 7c9878a into pytorch:main Feb 10, 2023
@pmeier pmeier deleted the cleanup-datapoints branch February 10, 2023 14:33
facebook-github-bot pushed a commit that referenced this pull request Mar 28, 2023
@NicolasHug @facebook-github-bot
[fbsync] remove datapoints compatibility for prototype datasets (#7154)
6fb095e 
Reviewed By: vmoens

Differential Revision: D44416260

fbshipit-source-id: 9437edffc8a7ccf08f381c5147d9a8f3e18530a3
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@NicolasHug NicolasHug NicolasHug approved these changes

Assignees
No one assigned
Labels
cla signed code quality module: datasets module: tv_tensors prototype
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
3 participants
@pmeier @NicolasHug @facebook-github-bot