[DEPRECATED] TFX on Cloud AI Platform Pipelines Reviews

Raul C. · Reviewed over 3 years ago

I did copy and past solutions, replacing endpoint and GCS uri Pipeline failed: .. beamapp-root-1220181845-6-12201018-dwoj-harness-6198 Root cause: Traceback (most recent call last): File "/usr/local/lib/python3.7/site-packages/apache_beam/runners/worker/sdk_worker.py", line 381, in get processor = self.cached_bundle_processors[bundle_descriptor_id].pop() IndexError: pop from empty list ..... File "/usr/local/lib/python3.7/site-packages/tensorflow_serving/apis/status_pb2.py", line 20, in <module> tensorflow_dot_tsl_dot_protobuf_dot_error__codes__pb2 = tensorflow_dot_core_dot_protobuf_dot_error__codes__pb2.tensorflow.tsl.protobuf.error_codes_pb2 AttributeError: module 'tensorflow.core.protobuf.error_codes_pb2' has no attribute 'tensorflow' time="2022-12-20T18:26:56.291Z" 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

Raimondo F. · Reviewed over 3 years ago

Error on 1st run of pipeline, maybe the problem is the trailing slash of cloud store uri. After fixing it, and deleting pipeline from kubeflow, recreating it only to get "pipeline already exist". recreate kubeflow instance and clean cs bucket, only to get again the same "pipeline already exist" and no pipeline in kf ui

Raimondo F. · Reviewed over 3 years ago

ok

Akhilesh S. · Reviewed over 3 years ago

pipeline transform block raises error : tensorflow_dot_tsl_dot_protobuf_dot_error__codes__pb2 = tensorflow_dot_core_dot_protobuf_dot_error__codes__pb2.tensorflow.tsl.protobuf.error_codes_pb2 AttributeError: module 'tensorflow.core.protobuf.error_codes_pb2' has no attribute 'tensorflow' time="2022-12-20T10:14:10.910Z" 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

Maxime P. · Reviewed over 3 years ago

Dave M. · Reviewed over 3 years ago

Too slow in creating a pipeline.

Srinivas K. · Reviewed over 3 years ago

Gabriel G. · Reviewed over 3 years ago

1. Deployment made and in kubeflow but not detected by the "check progress". 2. csvexamplegen task fails : 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 ... Something is missing somewhere

Carl-Erik G. · Reviewed over 3 years ago

Too slow in creating a pipeline.

Srinivas K. · Reviewed over 3 years ago

Checkers don't work. So I can't finish the course.

Aleh P. · Reviewed over 3 years ago

pipeline deployent didn't go through

Maxime P. · Reviewed over 3 years ago

Vijaya Satya Gowri Saikrishna D. · Reviewed over 3 years ago

Unable to get past the "Test completed tasks - Deploy the pipeline package to AI Platform Pipelines" even though the pipeline was properly deployed. This is my 2nd attempt for this very onerous lab. I am very very deeply disappointed as this may mean I have to stop and unenroll from the entire course, even though i wanted very badly to complete this

Steven S. · Reviewed over 3 years ago

too complicated.

Steven S. · Reviewed over 3 years ago

ok

Akhilesh S. · Reviewed over 3 years ago

Paitoon P. · Reviewed over 3 years ago

Vasanthan V. · Reviewed over 3 years ago

Nilay S. · Reviewed over 3 years ago

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

용기 지. · Reviewed over 3 years ago

Doesn't work

FIORELLA J. · Reviewed over 3 years ago

Konrad S. · Reviewed over 3 years ago

Amir O. · Reviewed over 3 years ago

Nicola M. · Reviewed over 3 years ago

Lab has an issue, you face issues when you get to run the pipline

Ibraheem T. · Reviewed over 3 years ago