[DEPRECATED] TFX on Cloud AI Platform Pipelines Reviews

trainer: 2023-12-18 14:31:06.913236: W tensorflow/stream_executor/platform/default/dso_loader.cc:59] Could not load dynamic library 'libcudart.so.10.1'; dlerror: libcudart.so.10.1: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/lo ...

dinh h. · Reviewed over 2 years ago

Teresa B. · Reviewed over 2 years ago

The cluster is not detected. I have created the cluster with the name and region specified in step 6 but the checkpoint does not detect the cluster. Despite I completed the lab, it is not marked as completed due to this bug

Gloria M. · Reviewed over 2 years ago

Danzler A. · Reviewed over 2 years ago

NEELESH S. · Reviewed over 2 years ago

Sandeep C. · Reviewed over 2 years ago

Alex L. · Reviewed over 2 years ago

blocked

Syed A. · Reviewed over 2 years ago

Lionel J. · Reviewed over 2 years ago

Damji S. · Reviewed over 2 years ago

At the end, the training pipeline failed due to errors with the cuda library and tensorflow

Fernando C. · Reviewed over 2 years ago

Giap L. · Reviewed over 2 years ago

Task1 instruction is outdated as pipeline moved to Vertex AI

harry z. · Reviewed over 2 years ago

JULIETTE M. · Reviewed over 2 years ago

notebook missing

Lionel J. · Reviewed over 2 years ago

notebook missing

Lionel J. · Reviewed over 2 years ago

Rohan D. · Reviewed over 2 years ago

SNEHANSU G. · Reviewed over 2 years ago

Азат И. · Reviewed over 2 years ago

Ashish T. · Reviewed over 2 years ago

Always failed the part of deploy AI

Ricardo I. · Reviewed over 2 years ago

Marie-Christine C. · Reviewed over 2 years ago

outdated

Julian Cæsar A. · Reviewed over 2 years ago

Utterly broken. As Freddie said, "You big disgrace"

Justin H. · Reviewed over 2 years ago

Good lab, but ends with error: pusher fails, detailed here: RuntimeError: Creating model verseion to AI Platform failed: <HttpError 400 when requesting https://ml.googleapis.com/v1/projects/qwiklabs-gcp-02-7e6af6099753/models/tfx_covertype_classifier/versions?alt=json returned "Field: version.deployment_uri Error: Deployment URI contains paths with empty directory names; please remove them and try again. Value specified: gs://qwiklabs-gcp-02-7e6af6099753-kubeflowpipelines-default//tfx_covertype_continuous_training/e9751bb9-14c9-4e83-85f8-d872e4f24ab7/Pusher/pushed_model/11/". Details: "[{'@type': 'type.googleapis.com/google.rpc.BadRequest', 'fieldViolations': [{'field': 'version.deployment_uri', 'description': 'Deployment URI contains paths with empty directory names; please remove them and try again. Value specified: gs://qwiklabs-gcp-02-7e6af6099753-kubeflowpipelines-default//tfx_covertype_continuous_training/e9751bb9-14c9-4e83-85f8-d872e4f24ab7/Pusher/pushed_model/11/'}]}]"> time="2023-12-12T23:57:09.702Z" level=info msg="sub-process exited" argo=true error="<nil>" time="2023-12-12T23:57:09.703Z" level=error msg="cannot save artifact /mlpipeline-ui-metadata.json" argo=true error="stat /mlpipeline-ui-metadata.json: no such file or directory" Error: exit status 1

Antoni N. · Reviewed over 2 years ago