Add the "tensorflow-model-analysis" package to the notebook images #544
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
More information about this package is [here](https://github.com/tensorflow/model-analysis).
|
/cc jlewi |
jlewi
reviewed
Mar 30, 2018
| @@ -206,6 +206,9 @@ RUN conda_packages=$(conda list -e | cut -d '=' -f 1 | grep -v '#' | sort) && \ | |||
| python -m ipykernel install --user && \ | |||
| echo "${pip_only_packages}" | xargs -n 1 -I "{}" /bin/bash -c 'pip install --no-cache-dir {} || true' && \ | |||
| pip install --no-cache-dir tensorflow-transform && \ | |||
| pip install --no-cache-dir tensorflow-model-analysis && \ | |||
| jupyter nbextension install --py --symlink tensorflow_model_analysis --system && \ | |||
There was a problem hiding this comment.
The reason I included it was that I started with the instructions from the model-analysis README which uses --symlink, and that part doesn't cause issues in our build (the default install location did cause issues, but that is orthogonal).
From what I understand, passing in --symlink is considered a best-practice since it saves space on non-Windows systems. However, I don't have a strong preference one way or the other.
|
/retest |
Installing the `tensorflow_model_analysis` nbextension results in the `tensorflow` package being imported. For the CPU image this is fine, but in the GPU image tensorflow relies on `libcuda`, which is not installed (it is expected to be mapped into the running container from the host OS). To work around this, we take the `libcuda.so` stub library (which is meant for building things that link against libcuda), and temporarily make it be loaded when something tries to load `libcuda.so.1`.
…arjur/tensorflow-model-analysis
|
Added a workaround to the issue where importing tensorflow during the build of the GPU image was failing. PTAL |
|
@ojarjur Could you add a comment in the Dockerfile explaining the work around? |
…arjur/tensorflow-model-analysis
…rflow-model-analysis extension
|
@jlewi Done. |
|
/lgtm |
|
[APPROVALNOTIFIER] This PR is APPROVED This pull-request has been approved by: jlewi The full list of commands accepted by this bot can be found here. The pull request process is described here
Needs approval from an approver in each of these files:
Approvers can indicate their approval by writing |
k8s-ci-robot
pushed a commit
that referenced
this pull request
Apr 3, 2018
* Update tf jupter notebook images to include tfma (from #544) (#570) * Update tf jupter notebook images to include tfma (from #544) tf1.4 gpu image has issues building with the latest tfma, so that's not updated in this PR /cc @jlewi /cc @ojarjur * Don't update tf images older than 1.6 * Moved OAuth secret from param to named secret (#572)
swiftdiaries
pushed a commit
to swiftdiaries/kubeflow
that referenced
this pull request
Apr 4, 2018
…ubeflow#544) * Add the "tensorflow-model-analysis" package to the notebook images More information about this package is [here](https://github.com/tensorflow/model-analysis). * Make the tensorflow library importable when building the GPU image. Installing the `tensorflow_model_analysis` nbextension results in the `tensorflow` package being imported. For the CPU image this is fine, but in the GPU image tensorflow relies on `libcuda`, which is not installed (it is expected to be mapped into the running container from the host OS). To work around this, we take the `libcuda.so` stub library (which is meant for building things that link against libcuda), and temporarily make it be loaded when something tries to load `libcuda.so.1`. * Document the LD_LIBRARY_PATH workaround used for installing the tensorflow-model-analysis extension
swiftdiaries
pushed a commit
to swiftdiaries/kubeflow
that referenced
this pull request
Apr 4, 2018
…ubeflow#570) * Update tf jupter notebook images to include tfma (from kubeflow#544) tf1.4 gpu image has issues building with the latest tfma, so that's not updated in this PR /cc @jlewi /cc @ojarjur * Don't update tf images older than 1.6
saffaalvi
pushed a commit
to StatCan/kubeflow
that referenced
this pull request
Feb 11, 2021
…ubeflow#544) * Add the "tensorflow-model-analysis" package to the notebook images More information about this package is [here](https://github.com/tensorflow/model-analysis). * Make the tensorflow library importable when building the GPU image. Installing the `tensorflow_model_analysis` nbextension results in the `tensorflow` package being imported. For the CPU image this is fine, but in the GPU image tensorflow relies on `libcuda`, which is not installed (it is expected to be mapped into the running container from the host OS). To work around this, we take the `libcuda.so` stub library (which is meant for building things that link against libcuda), and temporarily make it be loaded when something tries to load `libcuda.so.1`. * Document the LD_LIBRARY_PATH workaround used for installing the tensorflow-model-analysis extension
saffaalvi
pushed a commit
to StatCan/kubeflow
that referenced
this pull request
Feb 11, 2021
…ubeflow#570) * Update tf jupter notebook images to include tfma (from kubeflow#544) tf1.4 gpu image has issues building with the latest tfma, so that's not updated in this PR /cc @jlewi /cc @ojarjur * Don't update tf images older than 1.6
yanniszark
pushed a commit
to arrikto/kubeflow
that referenced
this pull request
Feb 15, 2021
* Ui changes trigger CI based on version * Make earlystop version in prow config
elenzio9
pushed a commit
to arrikto/kubeflow
that referenced
this pull request
Oct 31, 2022
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.
More information about this package is here.
This change is