Google Cloud Professional Machine Learning Engineer Cert Prep: 6 Monitoring, Optimizing, and Maintaining ML Solutions
1h 10mAdvanced2023-06-21
Authors
Noah Gift
MLOps Expert | Solopreneur | Author | Adjunct Professor | CTO
Course details
The Google Professional Machine Learning Engineer certification lets prospective employers know that you have the knowledge to design, build, and productionize machine learning models to solve business challenges using Google Cloud technologies, and knowledge of proven ML models and techniques.
In this sixth and final course in the series, Noah Gift covers monitoring, optimizing, and maintaining ML solutions. Noah starts with the key topic of data drift and its impact on model performance. He provides demos to ML solutions like auditing, logging, cloud web security scanners, and more. Noah also explains the “five whys” method of problem solving and why it’s an effective approach to debugging.
In this sixth and final course in the series, Noah Gift covers monitoring, optimizing, and maintaining ML solutions. Noah starts with the key topic of data drift and its impact on model performance. He provides demos to ML solutions like auditing, logging, cloud web security scanners, and more. Noah also explains the “five whys” method of problem solving and why it’s an effective approach to debugging.
Skills covered
Google CloudMachine LearningSoftware Development ToolsGoogleCloud PlatformsCert PrepArtificial Intelligence (AI)Cloud ComputingSoftware Development
Concepts
0. Introduction
- 01 - Overview
- 02 - Course six key terminology
1. ML Solutions
- 03 - Data drift explained by naughty child problem
- 04 - Load testing with Locust
- 05 - Demo - Auditing via logs
- 06 - Demo - Logging dashboard
- 07 - Demo - Cloud web security scanner
- 08 - Demo - Querying logging output with BigQuery
- 09 - Demo - Load testing with Rust
- 10 - Five whys
- 11 - Using Google Courses
- 12 - Building Rust HuggingFace Translator
- 13 - Using PyTorch Rust stable diffusion
- 14 - Using Rust with PyTorch
- 15 - Building a CUDA GPU stress test
Conclusion
- 16 - Next steps