GitLab

When running smoke test, proper way to install conda package is to
install from workspace, instead of installing from archive.
Otherwise the dependencies are not properly installed

https://github.com/conda/conda/issues/1884

For this, we pass the whole `conda-bld` directory from
conda package build job to upload job then smoke test job.

`torchaudio` does not have a mandatory third-party python package dependency,
so this was not an issue.

See also https://github.com/pytorch/text/pull/803

This CC adds CI test to check the backward compatibility of Torchscript functions/object dumps.

The job first dumps Torchscript objects into file from torchaudio 0.6.0 release environment (x Python 3.6, 3.7, 3,8), then load & run the function in master build (in Python 3.6, 3.7, 3.8).

If there is a BC-breaking change in master build, (registration schema change), then the test should fail.

At this moment, `info` function does not work due to suspected bug in torch side, so the test is disabled for `info` function. See https://github.com/pytorch/pytorch/issues/42258 for the detail. Once https://github.com/pytorch/pytorch/issues/42258 is resolved we can enable it.

* Use cmake for third party

* Apply patch to libmad

* Update gitignore

* Update docker test image

* packaging: Add test channels to pytorch dependency resolution
Signed-off-by: Eli Uriegas <eliuriegas@fb.com>

* .circleci: Add test channel to smoke tests
Signed-off-by: Eli Uriegas <eliuriegas@fb.com>

* .circleci: Put pytorch-test into a higher priority

pytorch-nightly was getting prioritized over pytorch-nightly which
shouldn't be the case
Signed-off-by: Eli Uriegas <eliuriegas@fb.com>

This PR refactors cache generation mechanism by introducing dedicated command
and bust cache on daily basis.

At this moment, Windows unittest job for 3.6 and 3.7 are broken because of
broken scipy but the environment is cached this persists until the next week.

As we have nightly build, we do not need to keep cache for one week.

* Add caching to binary build for more reliable build

* Apply same logic to unittest workflow

* Separate style check from unit test

* TEST flake8 fail

* TEST clang-format fail

* Revert "TEST flake8 fail"

This reverts commit 88d90185cefe294b9a662129a05c2e26a9844fe4.

* Revert "TEST clang-format fail"

This reverts commit 93517d1ccba650a6ccd813268fd73a4fd63b252c.

* Clean up extension build mechanism and extension location

* Add back the switch to depend on external sox

* Remove print

* Fix

1. Docker image used for smoke test is old (No Python 3.8). Updating this to the latest one. build [here](https://app.circleci.com/pipelines/github/pytorch/audio/1763/workflows/56c846a5-acaa-41a7-92f5-46ec66186c61/jobs/44484)
2. Add `latest` tag to the newly built docker so that later we can always use `latest` version and we do not have to update docker image tag manually.

* Add windows binary jobs

* Make wheel package platform-specific and python-version-specific

* Enable CUDA tests for Windows

* Add back branch filters

* Add GPU test

* Run only on master

* Turn on tests when building through conda-build

* skip one failling test on windows.

* Switch to unittest

* Fix conda env activation

* add linux only condition

* Fix librosa installation

* conda-forge channel as a fallback

* Revert "conda-forge channel as a fallback"

This reverts commit 8c15f6638e34792640a3de3e057d9c2995425180.

* Try the pytorch way

* Use conda-forge

* Disable conda-forge

* Install future at setup env

* Move linux unittest CI job tools to dedicated directory

* Restore Linux scripts and move the updated souces to a new subdir

* Add back future

* Remove comments
Co-authored-by: Vincent Quenneville-Belair <vincentqb@gmail.com>
Co-authored-by: moto <855818+mthrok@users.noreply.github.com>

* Bust cache weekly

* Rename directory for utility

* Show 20 slowest tests

* fixup! Show 20 slowest tests

* Add test image with Kaldi executable and third_parties

* Use Ubuntu as bese image since pytorch/many-linux is huge

* wip

* Add test

* add caching

* fixup! add caching

* fixup! fixup! add caching

pjh5 is a former employee so we should switch over to using a bot token.

Should resolve nightly pipeline failing on upload.
Signed-off-by: Eli Uriegas <eliuriegas@fb.com>

Signed-off-by: Eli Uriegas <eliuriegas@fb.com>

Python 2.7 was EOL on January 1, 2020

The last torchaudio release to support Python 2.7 was 0.4.0
Signed-off-by: Eli Uriegas <eliuriegas@fb.com>

* [WIP] Add Windows CI

* Remove cu_version

* checkout_merge -> checkout

* Add build script

* Switch backend to soundfile

* Remove soundfile as dependency

* Rename jobs

* Fix lint

Remove suspect logic.

Closes #304

See rationale writeup: https://github.com/pytorch/vision/pull/1321#issuecomment-531033978

* use binaries just built for smoke test

* to trigger tests on PR

* trigger tests

* different build has different test

* test good, now move to official

* test good, now move to official

* adding smoke test

* adding smoke test

* fix config.yml

* trigger smoke test

* test aws erc credential

* test aws erc credential

* test aws erc credential

* test aws erc credential

* adding ecr cred

* test aws erc credential

* test aws erc credential

* test aws erc credential - different way

* test aws erc credential - different way #2

* test aws erc credential - different way #3

* test aws erc credential - different way #4

* test aws #1

* test aws erc credential, after enabling circleci setting

* fixing conda init

* add verbose output for smoke test

* add docker image build workflow

* add docker image build workflow

* fix aws ecr login

* use images build from circelci

* ready for review

* updated Readme with instructions on nightly install, also updated docker image which can make nightlies installed via pip pass

* nit

* reduce size

Signed-off-by: Edward Z. Yang <ezyang@fb.com>