Introduction to MLOps with MLflow
summary
MLflow gets a hands‑on walkthrough: set up a tracking server with SQLite and local artifact root (mlruns), run a scikit‑learn DecisionTree on the penguins dataset, log parameters and test accuracy, and save the model artifact with a signature. The model becomes registered model v1 in the model registry, then is deployed locally via mlflow models serve to an HTTP /invocations endpoint and successfully returns predictions.
AI-written summary · about
chapters (17)
- 0:00Start
- 6:16Life Cycle Management
- 7:01Machine Learning Life Cycle
- 9:56Mlflow
- 11:42Model Registry
- 12:43Jupyter Notebooks
- 19:45Introduction to the Tracking Stores and the Artifact Stores
- 28:27Create an Experiment
- 46:23Test the Model
- 52:54Registering a Model
- 1:02:11Is It Possible To Lock the Entire Pipeline of the Model for Example the Pre-Processing Steps
- 1:02:36Custom Model
- 1:03:59How To Ensure that the Model Dependencies of the Pipeline Are the Same once We Serve the Model
- 1:05:19Write Your Custom Deployment Targets
- 1:07:31Tips and Tricks
- 1:09:17Is It Possible To Install the Dependencies in Runtime for Example if We Use Model the Model in an Api
- 1:18:04User and Permission Management
related talks
57mManaging Machine Learning Lifecycle with MLflow
31mMLflow: Platform for Complete Machine Learning Lifecycle
30mEnd-to-End MLOps with MLflow and Kubeflow
33mOperationalizing Model Lifecycle with MLFlow
45mMachine Learning Lifecycle Made Easy with MLflow
30mMLflow: an Open Source Tool for Managing the ML Project Lifecycle
24mMLflow: An open platform to simplify...
17mIntroducing mlflow