Leverage GitLab To Automate Machine Learning Pipelines
summary
Shows a working pipeline that builds, tests, reports on, and deploys a Rasa chatbot using Gitlab CI YAML and a Rasa dialog engine container image. Pipeline runs jobs to validate library version and training data, trains a model saved as an artifact, evaluates it on held‑out data, converts messy JSON metrics into a readable report and pushes it to the project wiki via Gitlab wiki API, then posts the model to production via server API; only one manual commit is required and the demo deploys successfully.
AI-written summary · about
related talks
24mFrom Model to Production: Automating the ML Lifecycle with CI/CD
46mFrom Dev to Prod with GitLab CI
24mReinvent your pipeline with GitLab, Kubernetes, and Amazon EKS
23mAccelerating Model Deployment with MLOps
52mFrom 0 to Continuous Deployments with Kubernetes, Helm and GitLab
51mGitLab CI/CD Pipelines with Kubernetes
31mAIOps: Automating and Scaling ML and LLM Workloads in Production
02h 47mPipes on Pipelines