ML MLOp Ops CI CI/CD CD for or Ma Machine Le Learn rning SASHA ROSENBAUM
Sasha Rosenbaum Sr. Program Manager @DivineOps
https://www.deliveryconf.com/
Agenda § Machine Learning 101 § ML CI/CD Pipeline Overview § Potential implementation § Demo
Trigger the pipeline!
What is MLOps and WHY should you care?
Machine Learning (ML) Is the science of getting computers to act Without being explicitly programmed
Machine Learning vs DevOps Google searches
Python questions on Stack Overflow
OK, but why should YOU care?
13
Data Scientists just want to Data Science
Deep Learning Some ML training algorithms are complex
Deep Learning - Backpropagation Some ML training algorithms are complex
Typical data scientist work environment
We’ve got the notebook into source control!
Programming Algorithm Answers Data
Machine Learning Answers Algorithm Data
Machine Learning Answers Model Data
Machine Learning Answers Model Data
Machine Learning Answers Data Model Predictions Data
How do we put the model in production?
What is an ML model?
Linear Regression – Housing Prices The training finds a and b such that Y = a+bX+ϵ
Deep Learning The input and output may be vectors ! 𝑌 , ! 𝑍
Image Classification
ML Model A definition of the mathematical formula with a number of parameters that are learned from the data
Isn’t this just an API endpoint?!
Do models really change that often?
Models must be improved continuously
The dataset matters!
The model predictions depend on what it has “seen”
=> Dataset is part of the model version!
TensorFlow Extended Te FB FBLear arne ner Fl Flow Ube Uber’s Michelangelo Mic Microso soft Ae Aether
But I don’t work at a big company with thousands of ML engineers!
How do we iterate?
Machine Learning Lifecycle
Data Scientist DevOps/SRE Quick iteration Quick iteration • • Versioning Versioning • • Reuse Reuse • • Great tools Compliance • • Ease of Observability • • Friends? management Uptime • Unlimited scale Updates • • Eliminating drift •
MLOps Workflow Collaborate Build app Test app Release app Monitor app App developer Data scientist Model reproducibility Model validation Model deployment Model retraining
MLOps Workflow Collaborate Build app Test app Release app Monitor app App developer Code, dataset, and environment versioning Data scientist Model reproducibility Model validation Model deployment Model retraining
MLOps Workflow Collaborate Build app Test app Release app Monitor app App developer Train model Automated ML ML Pipelines Hyperparameter tuning Data scientist Model reproducibility Model validation Model deployment Model retraining
MLOps Workflow Collaborate Build app Test app Release app Monitor app App developer Train model Validate model Model validation & certification Data scientist Model reproducibility Model validation Model deployment Model retraining
MLOps Workflow Collaborate Build app Test app Release app Monitor app App developer Train model Validate model Deploy model Model packaging Simple deployment Data scientist Model reproducibility Model validation Model deployment Model retraining
MLOps Workflow Collaborate Build app Test app Release app Monitor app App developer Train model Validate model Deploy Monitor model model Model management & monitoring Retrain model Data scientist Model performance analysis Model reproducibility Model validation Model deployment Model retraining
Build Your Own MLOps Platform + +
ML Pipeline A reusable, scaleable ML workflow template Kubeflow pipeline A reusable, scalable ML workflow template that runs on containers
Azure ML • Prep data • Train • Test • Deploy • Manage
Demo
Even a simple CI/CD pipeline is better than none!
DevOps Because change is the only constant in life
AI Ethics
Bias is a property of information
We must build AI responsibly
Build AI responsibly!
Thank You! @DivineOps
Questions?
Resources GitHub repo https://www.kubeflow.org/docs/azure/azureendtoend/ Deploy Kubeflow on Azure https://www.kubeflow.org/docs/azure/deploy/install-kubeflow/ Example Kubeflow Azure Pipeline https://www.kubeflow.org/docs/azure/azureendtoend/ Release pipeline https://dev.azure.com/sasrose/kubeflow/_release
Recommend
More recommend