Custom CUDA extensions #135
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
facebook-github-bot
added
the
CLA Signed
msaroufim
changed the title
cpuhrsch
approved these changes
Apr 15, 2024
weifengpy
reviewed
Apr 28, 2024
| @@ -62,5 +62,5 @@ jobs: | |||
| pip install ${{ matrix.torch-spec }} | |||
| pip install -r requirements.txt | |||
| pip install -r dev-requirements.txt | |||
| pip install . | |||
| python setup.py install | |||
There was a problem hiding this comment.
do we need to update README? for building from source, is it python setup.py developer now? I spent some time to understand why pip install -e . does not work anymore and find this PR
There was a problem hiding this comment.
argh, lemme update now
|
@msaroufim whether CPU kernel extension works with this PR? |
|
Yup cpu extensions work perfectly fine |
dbyoung18
pushed a commit
to dbyoung18/ao
that referenced
this pull request
Jul 31, 2024
* Custom CUDA extensionsCo-authored-by: Mark Saroufim <marksaroufim@meta.com> * ci fix * add continue on ci fail * add verbose pip logging * update * remove pip upgrade * remove torch from requirements * add python -m * readd upgrade * make everything python pip install * typo * bla * add torch again to dependencies * upd * udpate * one more * bla * uipda * bla * bla * add venv for pytest * skip test if no cuda * push * fix test * fix test * fix skip tests * reuse pip install . * add more debugging tools * yolo * revert yolo change * fix * remove custom install * print torch version * print torch version * teest' * try pip3? * remove pip3 * reshuffling torch install * no deps? * revert nodeps * ad * bla * bla * lazy * use conda and cuda toolkit * push * update * 1 more try * push * bla * bla * conda init * clowntown * this works locally * update * remove cache * push * alternate cuda install command * bla * bla * update * push * bla * bla * bla * bla * bla * bla * yolo * yolo * yolo * yolo * yolo * yolo * yolo * yolo --------- Co-authored-by: Richard Zou <rzou@meta.com>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This is the mergaeble version of #130 - some updates I have to make
conda install nvidia/label/cuda-12.1.0::cuda-toolkitto get things workingcontinue-on-error: trueto CI otherwise was difficult to debugpython setup.py installinsteadOK for manylinux I managed to repro the missing torch error locally (FINALLY) using this
Follow up PRs