Flexible ML Experiment Tracking System for Python Coders with DVC and Streamlit
summary
Claim: combine GIT plus DVC experiments via git refs and streamlit UI to load past experiment artifacts to get a lightweight, reproducible experiment tracker. Implementation: dvc repro / dvc exp run creates experiment commits, dvc.push stores outputs, and dvc.open reads model and prediction files from any experiment commit so the app can load and compare runs. Works: demo loads models, compares predictions, plots metrics. Caveats: quick to set up but requires ongoing code maintenance and careful Docker/CI handling for DVC remotes.
AI-written summary · about
chapters (16)
- 0:00Introduction
- 5:16Training Pipeline
- 12:00DVC Experiment
- 17:57Streamlit
- 24:50Whats next
- 25:38Python
- 28:53DVC
- 33:14Advantages and disadvantages
- 37:11Questions
- 37:53DVC and ML Flow
- 39:10GTB GDPR Compliance
- 39:31Presentation tool
- 40:48DVC advantages
- 42:18How does DVC experiments tracking work
- 43:13DVC infrastructure integration
- 44:03Conclusion
related talks
33mFlexible ML Experiment Tracking System for Python Coders with DVC and St...
27mLeveraging Git for ML Experiment Management
37mOpen Source Tools for ML Experiments Management
42mApplications in ML Drug Discovery pipelines
16mMake your data science workflows reproducible
01h 32mHow to easily set up and version control your Machine Learning Pipelines
02h 10mBuild End-To-End MLOps Platform with Open Source DVC Ecosystem
01h 25mBest practices and solution design using data versioning for machine learning