The SAME Project: A Cloud Native Approach to Reproducible Machine Learning
recording made private on YouTubehttps://youtu.be/aQvvGKVX25I
summary
SAME provides a one-line install and a CLI that bootstraps a local k3s cluster, deploys Kubeflow, and runs pipelines described in a declarative same file; it provisions services like Feast, TensorFlow, and Katib, imports eight historical runs into a metadata store, and continues run numbering so comparisons are immediate. Commands are same init and same run, pipelines run locally or against a cloud cluster, and a GitHub Action can tag and release models for reproducible production deployment.
AI-written summary · about
related talks
30mRealizing End to End Reproducible Machine Learning on Kubernetes
33mCompliant Data Management and Machine Learning on Kubernetes
54mScaling Machine Learning Pipelines in Cloud
51mMachine Learning using Kubeflow and Kubernetes
35mOrganizing data engineering around Kubeflow
15mKeynote: Machine Learning on Kubernetes Made Easy With Kubeflow
33mBuilding Reproducible ML Processes with an Open Source Stack
05mLightning Talk: Build a Reproducible ML Workflow with Kubeflow Pipelines