GCSB_MLE

This repository will be used to track and document my progress through the Google Cloud Skills Boost Machine Learning Engineer Learning Path. Each course in the learning path listed below is associated with an issue and a GitHub project is used to track overall progress. Work for each section is completed on a branch which is merged and closed upon completion.

Note: The section numbering below follows that given in the study guide where the first two introductory sections listed on the learning path page are not included in the numbering.

Learning path outline

Course 01. Introduction to AI and Machine Learning on Google Cloud (8 hours)

• Module 1: AI Foundations on Google Cloud
• Module 2: AI Development on Google Cloud
• Module 3: ML Workflow and Vertex AI
• Module 4: Generative AI on Google Cloud

Course 02. Prepare Data for ML APIs on Google Cloud (6.5 hours)

• Lab 1: Vertex AI: Qwik Start
• Lab 2: Dataprep: Qwik Start
• Lab 3: Dataflow: Qwik Start - Templates
• Lab 4: Dataflow: Qwik Start - Python
• Lab 5: Dataproc: Qwik Start - Console
• Lab 6: Dataproc: Qwik Start - Command Line
• Lab 7: Cloud Natural Language API: Qwik Start
• Lab 8: Speech-to-Text API: Qwik Start
• Lab 9: Video Intelligence: Qwik Start
• Lab 10: Prepare Data for ML APIs on Google Cloud: Challenge Lab

Course 03. Working with Notebooks in Vertex AI (0.75 hours)

Mini-course: 8 lessons

• Lesson 1: Working with Notebooks in Vertex AI
• Lesson 2: Vertex AI Notebook Solutions
• Lesson 3: Vertex AI Colab Enterprise notebooks
• Lesson 4: Vertex AI Workbench instance notebooks
• Summary
• Quiz: Working with Notebooks in Vertex AI
• Lab 1: Exploratory Data Analysis using Bigquery and Colab Enterprise (2 hrs)
• Lab 2: Exploratory Data Analysis using Bigquery and Workbench Instances (2 hrs)

Course 04. Create ML Models with BigQuery ML (5.5 hours)

• Lab 1: Getting Started with BigQuery ML
• Lab 2: Predict Visitor Purchases with a Classification Model in BigQuery ML
• Lab 3: Predict Taxi Fare with a BigQuery ML Forecasting Model
• Lab 4: Bracketology with Google Machine Learning
• Lab 5: Create ML Models with BigQuery ML: Challenge Lab

Course 05. Engineer Data for Predictive Modeling with BigQuery ML (4.25 hours)

• Lab 1: Creating a Data Transformation Pipeline with Cloud Dataprep
• Lab 2: ETL Processing on Google Cloud Using Dataflow and BigQuery (Python)
• Lab 3: Predict Visitor Purchases with a Classification Model in BigQuery ML
• Lab 4: Engineer Data for Predictive Modeling with BigQuery ML: Challenge Lab

Course 06. Feature Engineering (24 hours)

• Module 1: Introduction to Vertex AI Feature Store
• Module 2: Raw Data to Features
• Module 3: Feature Engineering
• Module 4: Preprocessing and Feature Creation
• Module 5: Feature Crosses: TensorFlow Playground
• Module 6: Introduction to TensorFlow Transform

Course 07. Build, Train and Deploy ML Models with Keras on Google Cloud (15.5 hours)

• Module 1: Introduction to the TensorFlow Ecosystem
• Module 2: Design and Build an Input Data Pipeline
• Module 3: Building Neural Networks with the TensorFlow and Keras API
• Module 4: Training at Scale with Vertex AI

Course 08. Production Machine Learning Systems (16 hours)

• Module 1: Architecting Production ML System
• Module 2: Designing Adaptable ML System Designing High-Performance ML Systems
• Module 3: Designing High-Performance ML Systems
• Module 4: Hybrid ML Systems
• Module 5: Troubleshooting ML Production Systems

Course 09. Machine Learning Operations (MLOps): Getting Started (8 hours)

• Module 1: Employing Machine Learning Operations
• Module 2: Vertex AI and MLOps on Vertex AI

Course 10. Machine Learning Operations (MLOps) with Vertex AI: Manage Features (8 hours)

• Module 1: Introduction to Vertex AI Feature Store
• Module 2: An In-Depth Look

Course 11. Introduction to Generative AI (0.75 hours)

• Mini-course: 1 lesson

Course 12. Introduction to Large Language Models (0.5 hours)

• Mini-course: 1 lesson

Course 13. Machine Learning Operations (MLOps) for Generative AI (0.5 hours)

• Mini Course: 5 lessons

Course 14. Machine Learning Operations (MLOps) with Vertex AI: Model Evaluation (2.5 hours)

• Module 1: Introduction to Model Evaluation
• Module 2: Model Evaluation for Generative AI

Course 15. ML Pipelines on Google Cloud (2.25 hours)

• Module 1: Introduction to TFX Pipelines
• Module 2: Pipeline Orchestration with TFX
• Module 3: Custom Components and CI/CD for TFX Pipelines
• Module 4: ML Metadata with TFX
• Module 5: Continuous Training with Multiple SDKs, KubeFlow & AI Platform Pipelines
• Module 6: Continuous Training with Cloud Composer
• Module 7: ML Pipelines with MLflow

Course 16. Build and Deploy Machine Learning Solutions on Vertex AI (8.25 hours)

• Lab 1: Vertex AI: Qwik Start
• Lab 2: Identify Damaged Car Parts with Vertex AutoML Vision
• Lab 3: Deploy a BigQuery ML Customer Churn Classifier to Vertex AI for Online Predictions
• Lab 4: Vertex Pipelines: Qwik Start
• Lab 5: Build and Deploy Machine Learning Solutions with Vertex AI: Challenge Lab

Course 17. Create Generative AI Applications on Google Cloud (4 hours)

• Module 1: Generative AI Applications
• Module 2: Prompts
• Module 3: Retrieval Augmented Generation (RAG)

Course 18. Responsible AI for Developers: Fairness and Bias (4 hours)

• Module 1: AI Interpretability and Transparency
• Module 2: Modernizing Infrastructure in the Cloud

Course 19. Responsible AI for Developers: Interpretability and Transparency (3 hours)

• Module 1: AI Interpretability and Transparency
• Module 2: Modernizing Infrastructure in the Cloud

Course 20. Responsible AI for Developers: Privacy and Safety (5 hours)

• Module 1: AI Privacy
• Module 2: AI Safety