Implementing Canary Releases of TensorFlow Model Deployments with Kubernetes and Cloud Service Mesh Reviews

Lakshmi Narayana J. · Reviewed about 3 years ago

Ivan C. · Reviewed about 3 years ago

please improve your instructions (eg.: open file xy and change ...)

Felix S. · Reviewed about 3 years ago

Task 6. "Creating TensorFlow Serving deployment of the ResNet50 model." did not finish deploying before time was up (>60minutes). Unfortunately leaving me unable to finish the lab

Koen E. · Reviewed about 3 years ago

Nam V. · Reviewed about 3 years ago

The lab is super slow. Faced the following issue at Task 6: Creating a kubeflow deployment takes indefinitely to become "READY" this is the last log after which I had to end the lab as only less than 10 mins were remaining to complete: ``` $ kubectl get deployments NAME READY UP-TO-DATE AVAILABLE AGE image-classifier-resnet50 0/1 1 0 25m ```

Mrinaal D. · Reviewed about 3 years ago

Amir M. · Reviewed about 3 years ago

The lab is super slow. Faced the following issue again: Creating a kubeflow deployment takes indefinitely to become "READY" this is the last log after which I had to end the lab as only less than 10 mins were remaining to complete: ``` $ kubectl get deployments NAME READY UP-TO-DATE AVAILABLE AGE image-classifier-resnet50 0/1 1 0 25m ```

Mrinaal D. · Reviewed about 3 years ago

The lab is super slow. 1. The step where the asmcli is installed, faced the following issue multiple times: ``` asmcli: Registering the cluster as cluster-1... asmcli: [WARNING]: Failed, retrying...(1 of 2) asmcli: [WARNING]: Failed, retrying...(2 of 2) asmcli: [WARNING]: Command 'run_command gcloud container hub memberships register cluster-1 --project=qwiklabs-gcp-01-659f7daa3394 --enable-workload-identity --gke-uri=https://container.googleapis.com/v1/projects/qwiklabs-gcp-01-659f7daa3394/zones/us-west1-c/clusters/cluster-1' failed. ``` Resolved this by refreshing the page after 5-6 times retrying the same command. 2. Creating a kubeflow deployment never became "READY" this is the last log after which I had to end the lab as only less than 10 mins were remaining to complete: ``` $ kubectl get deployments NAME READY UP-TO-DATE AVAILABLE AGE image-classifier-resnet50 0/1 1 0 25m ```

Mrinaal D. · Reviewed about 3 years ago

it's broken

Wouter D. · Reviewed about 3 years ago

Arash S. · Reviewed about 3 years ago

Aarón M. · Reviewed about 3 years ago

Paulina P. · Reviewed about 3 years ago

Miguel C. · Reviewed about 3 years ago

error not enough quota in task 2 while creating the cluster

Salma M. · Reviewed about 3 years ago

Lucas A. · Reviewed about 3 years ago

Ho N. · Reviewed about 3 years ago

Manish R. · Reviewed about 3 years ago

2 major problems 1. see this thread regarding insufficient quota error: https://community.deeplearning.ai/t/error-gcloud-beta-container-clusters-create-responseerror-code-403-message-insufficient-regional-quota-to-satisfy-request-resource-ssd-total-gb/291098 2. in task 7 the deployment is never ready, even when waiting for more than an hour. Therefore I couldn't finish the lab. Please fix this

Fran M. · Reviewed about 3 years ago

Manish R. · Reviewed about 3 years ago

navid k. · Reviewed about 3 years ago

Igor B. · Reviewed about 3 years ago

Danyail M. · Reviewed about 3 years ago

Freya K. · Reviewed about 3 years ago

doesnt work

Lucas M. · Reviewed about 3 years ago