Explore machine learning experiment tracking and model registry using the Neptune.ai platform in this comprehensive 53-minute tutorial. Learn how to log, store, query, display, organize, and compare model metadata in a single place. Discover key features that help you feel in control of model building, increase productivity in ML engineering and research, and build reproducible, compliant, and traceable models. Follow along with a hands-on heart disease detection ML experiment using Keras/TensorFlow, and see how to integrate Neptune tracking code. Dive into the Neptune ML Tracking Dashboard, compare experiments, add custom content tracking, and push your work to GitHub. Master essential MLOps skills to efficiently manage computational resources and streamline your machine learning workflow.
Smart ML Experiment Tracking and Model Registry with Neptune.AI Platform
Overview
Syllabus
- Video Start
- Introduction
- Why and Why?
- 3 Key Features
- Platform Service Status
- MLOps Tutorials
- Get Platform Access
- Jupyter Notebook colab & GitHub
- Heart Disease Detection ML Experiment
- ML Experiment in Keras/TensorFlow
- Export colab notebook to GitHub
- Adding Neptune Tracking code
- Create Tracking Project
- Initialize Neptune runtime objects
- ML Experiment with Neptune Objects
- Neptune Callback with ML Experiment
- Neptune ML Tracking Dashboard
- Adding another Experiment
- Comparing Experiments
- Custom content image Tracking
- More Experiments tracking
- Adding Jupyter notebook as resource
- Pushing Colab notebook to GitHub
- Recap
- Credits
Taught by
Prodramp
Related Courses
-
An AI Engineer Guide to Comet Model Registry Platform
-
An AI Engineer Guide to Model Monitoring with Comet ML Platform
-
Experiment Tracking in the Age of LLMs - MLOps Podcast
-
A Hitchhiker's Guide to the MLOps Experience
-
Kubeflow vs MLFlow: Choosing the Right MLOps Platform
-
Advanced Experiment Tracking for LLM-Powered Applications with Customized Open-Source MLFlow
