-
Notifications
You must be signed in to change notification settings - Fork 233
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.
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
Use torch for CUDA 11.8 in tests #2149
Conversation
Codecov Report
Additional details and impacted files@@ Coverage Diff @@
## develop #2149 +/- ##
===========================================
+ Coverage 36.23% 36.24% +0.01%
===========================================
Files 477 477
Lines 42544 42570 +26
===========================================
+ Hits 15414 15428 +14
- Misses 27130 27142 +12 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM
53da916
to
e8a9003
Compare
@@ -117,11 +117,11 @@ test-examples-tensorflow: | |||
# PyTorch backend | |||
install-torch-test: | |||
pip install -U pip | |||
pip install -e .[torch] | |||
pip install -r tests/torch/requirements.txt | |||
pip install -e .[torch] --index-url https://download.pytorch.org/whl/cu118 --extra-index-url=https://pypi.org/simple # ticket 119128 |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Tried to pip install -e .[torch] --extra-index-url=https://download.pytorch.org/whl/cu118
it's still install torch-2.0.1%2Bcu118-cp310-cp310-linux_x86_64.whl
.
Looks like it can be replaced pip install -e .[torch] --extra-index-url=https://download.pytorch.org/whl/cu118
With torch==2.0.1+cu118
bug is not reproduced 👍
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
That's what I tried too, but it didn't work in precommit. At the same time it works locally, maybe pip version is at fault. This is a temporary solution anyway until default torch starts shipping for CUDA 11.8.
No description provided.