Open Source Tools for ML Experiments Management
summary
Covers gaps between traditional Git/Git-LFS tooling and the needs of ML workflows—particularly data versioning, pipeline/versioning of experiments, metrics visualization and granular dataset/model artifact management—and motivates purpose-built tools. Describes the ideas behind the open-source tool DVC (dvc.org) as a concrete solution for managing ML experiments, large datasets, and models.
AI-written summary · about
chapters (8)
- 0:00Start
- 2:40101FFERENCE 1: HYPER PARAMETERS
- 8:34THE SPREADSHEET ISSUE: LACK OF TRUST
- 11:12MLFLOW TRACKING
- 21:18LOVE - ADO DATAFILE
- 23:05LOVE RETRIEVE AN EXPERIMENT
- 28:09THE WORLD IS CHANGING
- 28:45HOW TO DESIGN OUR FUTURE
related talks
27mLeveraging Git for ML Experiment Management
35mData versioning in machine learning projects
42mApplications in ML Drug Discovery pipelines
33mFlexible ML Experiment Tracking System for Python Coders with DVC and St...
59mDevOps for Machine Learning Projects
32mReproducibility and versioning of ML systems
46mFlexible ML Experiment Tracking System for Python Coders with DVC and Streamlit
02h 10mBuild End-To-End MLOps Platform with Open Source DVC Ecosystem