reconcile tensorflow serving version #409
Conversation
|
Hi @hougangliu. Thanks for your PR. I'm waiting for a kubeflow member to verify that this patch is reasonable to test. If it is, they should reply with Once the patch is verified, the new status will be reflected by the I understand the commands that are listed here. Instructions for interacting with me using PR comments are available here. If you have questions or suggestions related to my behavior, please file an issue against the kubernetes/test-infra repository. |
object_detection/tf_serving_cpu.md
Outdated
| @@ -17,12 +17,14 @@ MODEL_COMPONENT=pets-model | |||
| MODEL_PATH=/mnt/exported_graphs/saved_model | |||
| MODEL_STORAGE_TYPE=nfs | |||
| NFS_PVC_NAME=pets-pvc | |||
| TF_SERVING_VERSION=`kubectl -n kubeflow exec -ti tf-training-job-master-0 -- python -c 'from __future__ import print_function;import tensorflow;print(tensorflow.__version__, end="")'` | |||
There was a problem hiding this comment.
Is there a cleaner way to handle this? Why do we wait until runtime to get the TF version? Shouldn't this be determined by the code that gets checked in?
https://github.com/kubeflow/examples/blob/master/object_detection/docker/Dockerfile.training#L45
So can't we just have the code pin consistent version of TF across the example?
There was a problem hiding this comment.
Thanks for your comment. Agree with you. Since default OBJ_DETECTION_IMAGE tensorflow version is 1.10.0, we pin consistent version 1.10.0 of TF across the example.
Updated.
Since default OBJ_DETECTION_IMAGE tensorflow version is 1.10.0, we pin consistent version 1.10.0 of TF across the example. Fixes: kubeflow#408
hougangliu
force-pushed
the
reconcile_tf_serving_version
branch
from
85977ff to
9ce8a9f
Compare
|
/ok-to-test @hougangliu I think you should have recieved an invite to join the GitHub Kubeflow org; if you accept it that will allow you to trigger the tests. |
|
[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 |
Since default OBJ_DETECTION_IMAGE tensorflow version is 1.10.0, we pin consistent version 1.10.0 of TF across the example. Fixes: kubeflow#408
Fixes: #408
This change is