Build and Deploy Machine Learning Solutions on Vertex AI
This skill badge quest is for professional Data Scientists and Machine Learning Engineers. The datasets and labs are built around high business impact enterprise machine learning use cases; these include retail customer lifetime value prediction, mobile game churn prediction, visual car part defection identification, and fine tuning BERT for review sentiment classification. Learners who complete this skill badge will gain hands-on experience with Vertex AI for new and existing ML workloads and be able to leverage AutoML, custom training, and new MLOps services to significantly enhance development productivity and accelerate time to value.
A skill badge is an exclusive digital badge issued by Google Cloud in recognition of your proficiency with Google Cloud products and services and tests your ability to apply your knowledge in an interactive hands-on environment. Complete this Skill Badge, and the final assessment challenge lab, to receive a digital badge that you can share with your network.
When you complete this activity, you can earn the badge displayed above! View all the badges you have earned by visiting your profile page. Boost your cloud career by showing the world the skills you have developed!
In this lab, you will use BigQuery for data processing and exploratory data analysis, and the Vertex AI platform to train and deploy a custom TensorFlow Regressor model to predict customer lifetime value (CLV). The goal of the lab is to introduce to Vertex AI through a high value real world use case - predictive CLV. Starting with a local BigQuery and TensorFlow workflow, you will progress toward training and deploying your model in the cloud with Vertex AI.
In this lab, you will learn how to train a custom Vertex AI image classification model to recognize damaged car parts.
warning Deploy a BigQuery ML Customer Churn Classifier to Vertex AI for Online Predictions
In this lab, you will train, tune, evaluate, explain, and generate batch and online predictions with a BigQuery ML XGBoost model. You will use a Google Analytics 4 dataset from a real mobile application, Flood it!, to determine the likelihood of users returning to the application. You will generate batch predictions with your BigQuery ML model as well as export and deploy it to Vertex AI for online predictions.
In this lab you will create ML Pipelines using Vertex AI
In this challenge lab you will train, deploy, and create a model pipeline using Vertex AI.
- Write and train models locally in a hosted Vertex Notebook
- Containerize your training code and push it to Google Cloud Artifact Registry
- Create managed image dataset artifacts for experiment tracking
- Train a BigQuery ML (BQML) XGBoost classifier
- Trigger a training job using the Vertex AI Console
- Create a custom model evaluation component using the KFP SDK
- Incorporate pre-built KFP components into your pipeline
- Deploy your trained models to a Vertex Endpoint for online predictions
- Query your model for online predictions and explanations