How to get predictions from an ML model

Опубликовано 25 августа 2022, 16:00

Welcome back for another episode of Prototype to Production! In the previous video, Developer Advocate Nikita Namjoshi trained a custom model on Vertex AI. In this video, Nikita discusses how to get batch and online predictions from your custom-trained model. Watch along and learn how to solve real-world problems with machine learning on Vertex AI.

Chapters:
0:00 - Intro
0:24 - Model predictions in notebook
1:07 - How to import your ML model into Vertex AI
2:14 - What are batch and online predictions?
2:57 - How to deploy your model
3:36 - Testing the model from a Workbench instance
5:53 - Wrap up

Getting predictions on Vertex AI codelab → goo.gle/3Al7rZN
Serving TensorFlow image models notebook → goo.gle/3dYO95a
Custom prediction routines → goo.gle/3R0IaLq
Pre-built containers for predictions → goo.gle/3CuQDCy
Custom containers for predictions → goo.gle/3cgFvhI

Prototype To Production playlist → goo.gle/PrototypeToProduction
Subscribe to Google Cloud Tech → goo.gle/GoogleCloudTech
​​
#Prototype2Production