MLOps Automation From A to Z | Jupyter + KubeFlow + MLRun + Nuclio

preview_player
Показать описание
0:00 - Intro to MLOps
0:43 - MLOps Tasks
1:12 - ML Pipeline
2:37 - Jupyter Notebook
3:56 - Serverless Function Marketplace
5:03 - MLRun UI
5:52 - Automated KubeFlow Pipeline
8:27 - KubeFlow Graph
10:44 - Inferencing API Canary Deployment
11:45 - Monitor & Visualize (Grafana)
12:47 - Summary
#presto #hive #airflow #mlrun #nuclio #kubernetes #k8s #kubeflow #mlflow #machinelearning #mlops #grafana #prometheus #horovod #dask #pandas #scikit #sklearn #python #r #datascience #deeplearning #xgboost #rapids #api #cluster #artificialintelligence
Рекомендации по теме
Комментарии
Автор

I'm sure it wouldn't be hard to customize, but just curious -- does the canary deployment system here support multiple canary models or just one at a time?

MrSupergingerman
Автор

HI, This is pretty informative .. I have a question how we have to implement MLrun with all this?

ashishjain
Автор

Too much technical terms. Can you do a detailed video on mlrun with an example

prajnasbhat