GitLab

* _weight_size -> _file_size

* Better formatting of individual weights tables

* Remove file size from main tables to avoid confusion with weight size (as in RAM)

* Remove unnecessary (file size) suffix

* Fix CI error?

* Formatting
Co-authored-by: Philip Meier <github.pmeier@posteo.de>

* pin python version for macos cmake

* ping python for all OS

* revert to only set this for macos

* debug

* remove debug

* [WIP] Remove torch.jit.fuser("fuser2") in test

Internally we're considering removing support for fuser2, so we need to remove this special case from the test.

* completely remove special-casing

* np.int -> np.int32

* Update test/test_transforms.py

[FBcode->GH]
https://www.internalfb.com/diff/D41268423

Co-authored-by: Samantha Andow <samdow@meta.com>

add to python script

Add cuda118

Add cuda 11.8 constrain

* add moving mnist dataset
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>

* remove unused modules
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>

* modify docstring
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>

* modify docstring and docs
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>

* add split and split ratio kwargs
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>

* fix checking split argument
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>

* remove unused package

* delete lines
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>

* fix filename property
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>

* fix reviews
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>

* modify docstrings
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>

* add split tests and etc
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>

* fix tests
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>
Signed-off-by: tsugumi-sys <tidemark0105@gmail.com>

* style: Added missing typing annotations in datasets/_stereo_matching

* style: Specified typing

* style: Specified type annotations further

* style: Specified typing of __getitem__

* style: Fixed linting
Co-authored-by: Philip Meier <github.pmeier@posteo.de>

* improve prototype GPU CI config

* consolidate CPU and GPU prototype workflows

* [REVERTME] deactivate other CI

* syntax

* fix name

* fix echo mutex

* style

* fix runner

* add jobname

* run all prototype tests

* fix mutex echo

* cleanup

* Revert "[REVERTME] deactivate other CI"

This reverts commit c931ff2c27bbc2b04324b777ae39a6fc25e50b8a.

* remove old workflows

* use free GHA runner for CPU tests

* Revert "use free GHA runner for CPU tests"

This reverts commit a209b31387d9d14bc6446743587db244bccc2106.

* [REVERTME] turn warnings into errors

* try fix test warnings

* fix module import in tests

* revert

* int -> float

* [REVERTME] why do we have to mark as safe?

Co-authored-by: Philip Meier <github.pmeier@posteo.de>

* style: Added typing annotations to datasets/_optical_flow

* style: Reverted back to str typing
Co-authored-by: Philip Meier <github.pmeier@posteo.de>

* Try add GPU ci for prototype transforms

* Update prototype-transforms-tests-linux-gpu.yml

* Update prototype-transforms-tests-linux-gpu.yml

* Update prototype-transforms-tests-linux-gpu.yml

* Update prototype-transforms-tests-linux-gpu.yml

* Update prototype-transforms-tests-linux-gpu.yml

* add small tolerance for adjust_contrast on CUDA

* [skip-ci] update
Co-authored-by: Philip Meier <github.pmeier@posteo.de>

* Add cuda resnet50 test

* Fix path

* Tune vision smoke test

* Add more advanced smoke test

* add torch import

* remove dependency on torch

* Add missing vars

* More code and ufmt
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* improve download tests

* move SBU to xfailed

* add SSL version output to CI

* fix prototype dataset imagenet mock data

* [REVERT ME] temporarily add prototype datasets test to CI

* Revert "[REVERT ME] temporarily add prototype datasets test to CI"

This reverts commit 7cafb5f01487c7625bcffd88e76832349b3e5055.

* rename features._Feature to datapoints.Datapoint

* _Datapoint to Datapoint

* move is_simple_tensor to transforms.utils

* fix CI

* move Datapoint out of public namespace

* Update to ROCm 5.3

* Regenerate config.yml

Co-authored-by: Philip Meier <github.pmeier@posteo.de>

* Adding _ops and _weight_size metadata checks to tests

* Fixing wrong ops value

* Changing test_schema_meta_validation to instantiate the model only once

* moving instantiating quantized models inside get_ops

* Small refactor of test_schema_meta_validation logic

* Reverting to previous ops value

* Simplifying unquantized models logic in test_schema_meta_validation

* Update test/test_extended_models.py
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>
Co-authored-by: Toni Blaslov <tblaslov@fb.com>
Co-authored-by: Vasilis Vryniotis <datumbox@users.noreply.github.com>

* expose some prototype transforms utils

* rename _isinstance

* [Nova] Disable CircleCI Linux Unittests

* add back linux gpu jobs

* Fix extensions in wheels build

* pass pre-script from caller

* add pre-script to run relocate on wheel

* install auditwheel for relocate

* reorder numpy install

* use jpeg80 for py3.9 and py3.10 mac builds

* revert back to jpeg9

* printing the found jpeg libs during build time and reintro circle for comparison

* more debug logs, remove extra linux builds

* typo fix

* brew uninstall conflicting jpeg libraries on mac machines

* fix path for unittests

* ufmt

* introduce type failures in dispatchers

* add type checks to all dispatchers

* add missing else

* add test

* fix convert_color_space

* minor cleanup of the prototype transforms tests

* refactor ImagePair

* pretty format enum

* Remove `_FT` usages from functional

* Update error messages

* use non-random images for interpolation kernels for testing

* use real image rather than artificial

* cleanup