End-To-End: Pipeline Orchestration (KFP) - AutoML in Vertex AI for ML Operations [notebook 02c]

An end-to-end workflow using Pipelines within Vertex AI on Google Cloud Platform. We will use AutoML to train a machine learning model. A walkthrough of building a repeatable pipeline to orchestrate all the steps from connecting to data sources, training a model, evaluating the final model, deploying to an online endpoint and requesting predictions from multiple clients. A few deep dives along the way including model explainability! This video follows the notebook 02c - Vertex AI - Pipelines - AutoML with clients (code) in an automated pipeline.

GitHub Repository: https://github.com/statmike/vertex-ai-mlops

The Notebook followed in this video:
https://github.com/statmike/vertex-ai-mlops/blob/main/02%20-%20Vertex%20AI%20AutoML/02c%20-%20Vertex%20AI%20%3E%20Pipelines%20-%20AutoML%20with%20clients%20(code)%20In%20automated%20pipeline.ipynb

Timeline:
0:00 - Introduction
1:00 - Overview
3:50 - Start Walkthrough
5:36 - [Notebook Section] Setup
10:30 - [Notebook Section] Pipeline Definition (part 1)
11:13 - Q&A: AutoML model types?
16:08 - [Notebook Section] Pipeline Definition (part 2)
20:27 - Q&A: What optimization objective?
23:36 - [Notebook Section] Pipeline Definition (part 3)
26:38 - [Notebook Section] Compile Pipeline
28:00 - [Notebook Section] Create Vertex AI Pipeline Job
30:40 - Review Completed Pipeline
35:40 - [Notebook Section] Prediction
44:57 - [Notebook Section] Batch Prediction
47:00 - [Notebook Section] Explanations (part 1)
47:17 - Q&A: What are explanations?
50:50 - [Notebook Section] Explanations (part 2)
53:21 - Q&A: When should I use pipelines for AutoML?
56:04 - Wrap-up

Скачать видео