ccd-ahm-2022

Contains my code and deck for Cloud Community Days Ahmedabad 2022: Fantastic ML Deployments and How to Do Them with Vertex AI.

Prerequisites

The notebooks provided in this repository use paid Google Cloud Platform (GCP) services:

• Vertex AI Workbench
• Google Cloud Storage (GCS)
• Vertex AI Prediction
• Google Container Registry (GCR)

So, I assume you already have a billing-enabled GCP Project and you can:

• spin up Vertex AI Workbench instances
• create buckets on GCS
• push Docker images to GCR

Notebooks

The notebooks are in notebooks directory. Below is a brief description of each notebook:

• resnet-export.ipynb shows how to export a TensorFlow model (image recognition) compatible with off-the-shelf Vertex AI deployment. The deployment part will be covered via the Vertex AI GUI console.
• onnx-conversion.ipynb shows how to optimized the TensorFlow model with ONNX and compares the latency of the both the models (TensorFlow and ONNX).
• custom-onnx-deployment.ipynb shows how to deploy the ONNX model to Vertex AI using custom prediction routes.

Slides

Link

Acknowledgements

Thanks to the ML Developer Programs team at Google for providing GCP credit support.