-
Notifications
You must be signed in to change notification settings - Fork 966
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
Upgrade pyarrow #1317
Upgrade pyarrow #1317
Conversation
http://b/310037300
This could be a problem:
|
Honestly, I think we can live with that. Very few metrics use GPUs and most of those are processing model files and so won't be affected by PyArrow versions one way or the other. There should be so few remaining GPU metrics that we can reasonably just plan on not enabling parquet support for them. |
Dockerfile.tmpl
Outdated
@@ -114,7 +114,7 @@ RUN pip install spacy && \ | |||
|
|||
# b/232247930: uninstall pyarrow to avoid double installation with the GPU specific version. | |||
RUN pip uninstall -y pyarrow && \ | |||
mamba install -y cudf cuml && \ | |||
mamba install -y "pyarrow>=14.0.1" cudf cuml && \ |
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.
Let's see if this passes, if it does we'll upgrade both, if not, then we'll switch to only upgrading the CPU-image.
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.
Note that without this change, it does look like tests pass:
https://ci.kaggle.net/blue/organizations/jenkins/kernels%2Fdocker-python/detail/upgrade-pyarrow/1/pipeline/58/
So cudf is complaining, but it might work anyway...? Though we might not want to risk that.
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.
I'd prefer not to risk it.
http://b/310037300