Using custom containers with AI Platform Training Reviews

The notebook crashed and I was unable to complete the final steps.

Cataron, A. · Reviewed over 4 years ago

Ewan Z. · Reviewed over 4 years ago

GIORGOS S. · Reviewed over 4 years ago

Satimbayev M. · Reviewed over 4 years ago

Vasco F. · Reviewed over 4 years ago

Rodrigo S. · Reviewed over 4 years ago

Eddy G. · Reviewed over 4 years ago

Eddy G. · Reviewed over 4 years ago

Eddy G. · Reviewed over 4 years ago

Estefania L. · Reviewed over 4 years ago

Antonio M. · Reviewed over 4 years ago

Dzmitry H. · Reviewed over 4 years ago

Pavan P. · Reviewed over 4 years ago

Notebooks no longer in Vertex, incorrect commands when generating cluster (specified version not found, had to upgrade to .19 and remove basic auth)

Nick H. · Reviewed over 4 years ago

Tan K. · Reviewed over 4 years ago

Efthymios S. · Reviewed over 4 years ago

The more I look at this lab, the more I think it's really, really badly designed. It seems that many students are giving up early because they can't create a pipeline, as the lab doesn't come provisioned with a cluster that we're told is necessary for the pipeline. But to the best of my understanding, this pipeline is never even used, since the entirety of this lab is run from a Jupyterlab notebook. If someone at Qwiklabs who actually understood ML pipelines took a look at this lab and got rid of the superfluous steps, a LOT of unnecessary frustration could be avoided.

Matthew J. · Reviewed over 4 years ago

This lab is completely outdated

Sheiss H. · Reviewed over 4 years ago

Very frustrated. This lab is filled with problems that have already been listed by other students, but there's no sign of any of these problems being addressed. Also, two hours is not enough time for this lab. Building images, creating models, etc... takes a considerable amount of time, and a single problem is even to cause a student to run out of time.

Matthew J. · Reviewed over 4 years ago

Ran twice and didn't finish. The AI Platform is slow or unstable even for such a toy model hyperparameter tuning.

Jian C. · Reviewed over 4 years ago

Ekaterina P. · Reviewed over 4 years ago

Couldnt create the kubernetes cluster! ERROR: (gcloud.container.clusters.create) ResponseError: code=400, message=Basic authentication was removed for GKE cluster versions >= 1.19. The cluster cannot be created with basic authentication enabled. Instructions for choosing an alternative authentication method can be found at: https://cloud.google.com/kubernetes-engine/docs/how-to/api-server-authentication.

Michel A. · Reviewed over 4 years ago

Brandon G. · Reviewed over 4 years ago

Bas G. · Reviewed over 4 years ago

Tan K. · Reviewed over 4 years ago