Save model to Vertex AI

Course Q&A Machine Learning Engineering for Production(MLOps) Deploying Machine Learning Models in Production
1

Course practice example shows how to save a model to AI Platform. However, now GCP uses Vertex AI instead. So, I used Vertex AI workbench to create a Jupyter lab notebook. And, saved a model to a bucket. Then, how can I save a model to Vertex AI, not to AI Platform?

!gcloud ai-platform versions create $VERSION_NAME

→ should it be something like…below?

!gcloud vertex-ai versions create $VERSION_NAME

I am getting stuck at creating a model version:

vertex_ai
vertex_ai964×216 85.8 KB

I followed this course Notebook:

2

Hello @Jaelin_Lee and Welcome to DeepLearning.AI
If I understood your question right I think you’re asking how to deploy an AI model using Vertex AI and not an AI platform.

First is good to note: ‘Vertex AI’ is also an ‘AI platform’ below are some suggestion

Deploying the model

Deploying the model is a three-step process. It can be performed with the API, gcloud, or the available SDKs for different programming languages.

  1. You have to upload your model to the Vertex AI Model Registry.
  2. Create a Vertex AI Endpoint (once)
  3. After that, you can deploy your model to the Vertex AI Endpoint.

Please refer to this article for more information Serving Machine Learning models with Google Vertex AI

Happy Learning
Isaak Kamau