Skip to content

Conversation

kazhang
Copy link
Contributor

@kazhang kazhang commented Nov 29, 2021

Log IO api usages.
Partially-close: #4955

@facebook-github-bot
Copy link
Contributor

facebook-github-bot commented Nov 29, 2021

💊 CI failures summary and remediations

As of commit c78335e (more details on the Dr. CI page):


💚 💚 Looks good so far! There are no failures yet. 💚 💚


This comment was automatically generated by Dr. CI (expand for details).

Please report bugs/suggestions to the (internal) Dr. CI Users group.

Click here to manually regenerate this comment.

@kazhang kazhang requested a review from fmassa November 29, 2021 23:43
@kazhang kazhang marked this pull request as ready for review November 30, 2021 07:06
* [DIRTY] cleanup CI config

* remove pip install command

* fix syntax

* fix jobs

* fix syntax

* add support for editable install

* sync

* fix sudo apt install

* fix editable install

* sync

* try pip without user install

* [DIRTY] address review comments

* [DIRTY] try use dynamic name

* switch logic

* address remaining comments

* cleanup

* more cleanup

* fix enabling prototype tests

* Update .circleci/config.yml.in

Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

* linebreak

Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>
Copy link
Contributor

@datumbox datumbox left a comment

Choose a reason for hiding this comment

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

Thanks for the PR @kazhang.

The added logging looks good. Some of the calls (for example read_image()) will generate additional logging as they call other public methods but I don't think this is an issue or something we can avoid.

There is at least one more logging needed at the constructor of VideoReader defined here.

Finally, does it make sense to consider at this point adding the logging calls on the C++ side so that we can include cases where the Python interface is not used? I think that has been discussed offline at one point.

datumbox and others added 16 commits November 30, 2021 12:56
* Change the `default` weights mechanism to sue Enum aliases.

* Change `get_weights` to work with full Enum names and make it public.

* Applying improvements from code review.
* improve COCO prototype

* test 2017 annotations

* add option to include captions

* fix categories and add tests

* cleanup

* add correct image size to bounding boxes

* fix annotation collation

* appease mypy

* add benchmark

* always use image as reference

* another refactor

* add support for segmentations

* add support for segmentations

* fix CI dependencies
making torchvision ops leaf nodes by default
Co-authored-by: Prabhat Roy <prabhatroy@fb.com>
* Add bias parameter to ConvNormActivation

* Update torchvision/ops/misc.py

Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>
According to the benchmark link #1311 (comment), for GPU the threshold should be higher (which is why detectron2 used 40k).
This PR changes the threshold to be device dependent.

Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>
Co-authored-by: Francisco Massa <fvsmassa@gmail.com>
* support amp training for video classification models

* Removed extra empty line and used scaler instead of args.amp as function argument

* apply formating to pass lint tests

Co-authored-by: Konstantinos Bozas <kbz@kbz-mbp.broadband>
* exclude windows from mmap

* add comment for windows

* appease mypy
Copy link
Contributor

@datumbox datumbox left a comment

Choose a reason for hiding this comment

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

I have a few questions up for discussion, other than that the changes look good to me.

const torch::Tensor& data,
ImageReadMode mode,
bool allow_16_bits) {
C10_LOG_API_USAGE_ONCE("torchvision.io.decode_png");
Copy link
Contributor

Choose a reason for hiding this comment

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

Same here and in all other places where the string on the C++ side matches the one on Python.

Copy link
Contributor

@prabhat00155 prabhat00155 left a comment

Choose a reason for hiding this comment

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

Thanks @kazhang!

Borda and others added 4 commits December 6, 2021 17:01
* adding PR template

* update

* Update PULL_REQUEST_TEMPLATE.md

* Update PULL_REQUEST_TEMPLATE.md

* Update .github/PULL_REQUEST_TEMPLATE.md

Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>

Co-authored-by: Nicolas Hug <contact@nicolas-hug.com>
Copy link
Contributor

@datumbox datumbox left a comment

Choose a reason for hiding this comment

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

@kazhang It seems that something is wrong with the PR: 78 modified files and 3.1k additions/1.6k deletions.

I haven't checked which commit caused this but perhaps its worth force pushing to undo or starting a new one. Marking as "request changes" to avoid accidental merges while the PR is in this state.

@kazhang kazhang closed this Dec 7, 2021
@kazhang
Copy link
Contributor Author

kazhang commented Dec 7, 2021

@datumbox I created a new PR: #5038

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Add _log_api_usage_once() calls on remaining endpoints

10 participants