-
Notifications
You must be signed in to change notification settings - Fork 69
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
Metrics changes #286
Metrics changes #286
Conversation
Build is failing with an error: Please replace these lines with following:
|
docker/1.5-1/final/Dockerfile.cpu
Outdated
@@ -1,4 +1,4 @@ | |||
ARG SAGEMAKER_XGBOOST_VERSION=1.5-1 | |||
iARG SAGEMAKER_XGBOOST_VERSION=1.5-1 |
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.
nit should be ARG
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.
Oh oops, sorry about that!
docker/1.5-1/final/Dockerfile.cpu
Outdated
@@ -15,7 +15,9 @@ RUN python3 -m pip install -r /requirements.txt && rm /requirements.txt | |||
# Copy wheel to container # | |||
########################### | |||
COPY dist/sagemaker_xgboost_container-2.0-py2.py3-none-any.whl /sagemaker_xgboost_container-1.0-py2.py3-none-any.whl | |||
RUN rm -rf /miniconda3/lib/python3.7/site-packages/numpy-1.21.2.dist-info && \ | |||
|
|||
RUN conda install numpy --force-reinstall && \ |
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.
Why do we force-reinstall
the numpy here? Will it install the latest numpy version? Shouldn't we pin https://github.com/aws/sagemaker-xgboost-container/blob/master/requirements.txt#L11 here and use pinned numpy?
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.
The force-reinstall was due to an earlier comment. I did try running it without and there is a build error associated with numpy. I believe the reinstall will reinstall the current version and not the latest version but I will confirm this is the case when the check finishes.
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.
Can you please try to update numpy version to numpy==1.21.6
in requirements.txt and check if this change can be avoided?
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.
The change to the numpy version has been added!
Integration Tests for the new metrics (MAE and RMSE): [mxnet-algorithms-integ-test-220601-191722-724128] |
5abd69d
to
2552292
Compare
Co-authored-by: Mark Bunday <mabunday@amazon.com>
* Bump docker runtime version from 17 to 19 * Remove explicit requests install from buildspec Co-authored-by: Mark Bunday <mabunday@amazon.com>
Can you please confirm if the integ tests were run for an image containing these changes? |
Yes, the integration testing was ran on a docker image with the new changes in this PR |
Issue #, if available:
Description of changes:
Adding new metrics and tests
By submitting this pull request, I confirm that you can use, modify, copy, and redistribute this contribution, under the terms of your choice.