TFX on Google Cloud Vertex AI Pipelines Reviews

Wulan D. · Reviewed over 2 years ago

Slamet Rofikoh C. · Reviewed over 2 years ago

installation instructions for tf and tfx are outdated

Anton K. · Reviewed over 2 years ago

Latest versions of tensorflow/tfx have a dependency issue with shapely, it took me a while to fix it.. A simple workaround would be to add "!pip uninstall shapely tensorflow tensorflow-io tensorboard -y" before the pip upgrade.

Aghiles A. · Reviewed over 2 years ago

Oliver F. · Reviewed over 2 years ago

Wulan D. · Reviewed over 2 years ago

Al-Fariqy R. · Reviewed over 2 years ago

Wulan D. · Reviewed over 2 years ago

Completely broken notebook! I hope I won't be charged for another attempt!

Haim E. · Reviewed over 2 years ago

Ricardo Z. · Reviewed over 2 years ago

Wulan D. · Reviewed over 2 years ago

the notebook did not work - library specification is incorrect. I spent the whole lab trying to get the right version of tfx and kfp installed to run a pipeline, but could not. I created an issue in the github repository to alert the notebook creators about this.

Jason B. · Reviewed over 2 years ago

Lot of errors

Sabyasachi M. · Reviewed over 2 years ago

Error TFX on this lab. Please fix that

Muhammad Rafli P. · Reviewed over 2 years ago

the vertex notebook does not work properly

Guillermo A. · Reviewed over 2 years ago

Dmitrii P. · Reviewed over 2 years ago

Anrie S. · Reviewed over 2 years ago

Please fix this lab, theres some error in jupyter

Muhammad Rafli P. · Reviewed over 2 years ago

Annisa M. · Reviewed over 2 years ago

Bug error at jupyter lab

Muhammad Rafli P. · Reviewed over 2 years ago

Galih T. · Reviewed over 2 years ago

Had to fight for hours to have jupyterlab to import tfx succesfully.

Lalli M. · Reviewed over 2 years ago

Took bit too long and installing of library/restart runtime had issues so I had to run the same step several times

Jung S. · Reviewed over 2 years ago

Package installation is broken. The first step in the notebook is to upgrade pip and install packages, and it fails. I was able to make it work by removing the pip upgrade part, but even in that case there were issues with installing kfp and dependency resolving. Also it may be useful to add note about switching regions in the Vertex Pipelines UI - I was confused when I did not see my pipeline, but the problem was that the default region was no the one where the run was created.

Kirill D. · Reviewed over 2 years ago

Old Code!

Arjun K. · Reviewed over 2 years ago