Vertex Forecasting and Time Series in Practice

Module 1: Course Introduction

• Identify the reasons to learn Vertex AI Forecasting from Google.

• Learn the course objectives.

Module 2: Time Series and Forecasting Fundamentals

• Identify the different types of sequence models.

• Identify the different patterns and analysis methods of time series.

• Describe the primary notations of forecasting.

Module 3: Forecasting Options on Google Cloud

• Identify the options to develop forecasting models on Google Cloud.

• Describe Vertex AI and its benefits.

• Explore the workflow to build a forecasting model by using Vertex AI.

Module 4: Data Preparation

• Prepare the input data to fit the requirements of Vertex AI Forecasting.

• Demonstrate different types of features.

• Describe the best practices for the data ingestion stage.

Module 5: Model Training

• Configure model training.

• Select the appropriate training optimization objective.

Module 6: Model Evaluation

• Demonstrate training data split in time series forecasting.

• Describe evaluation metrics.

Design the approach to improve the performance.
Module 7: Model Deployment

• Deploy the forecasting model.

• Describe Vertex AI Pipelines and MLOps

• Use batch predictions to generate model forecasts.

Module 8: Model Monitoring

• Describe model drift.

• Demonstrate model retraining.

• Use Vertex AI Pipelines and prebuilt (SDKs) to automate the forecasting workflow.

Module 9: Vertex Forecasting in Retail

• Describe the steps and considerations of building a forecasting solution in retail.

• Demonstrate the model development with different datasets.

• Identify the challenges and the lessons of developing a forecasting model in retail.
  

Professional data analysts, data scientists, and ML engineers who want to build end-to-end high performance forecasting solutions on Google Cloud and add automation to the workflow.

Prerequisite:

Having one or more of the following:

• Basic knowledge of Python syntax

• Basic understanding of machine learning models

• Prior experience building machine learning solutions on Google Cloud