Transforming a Jupyter Notebook into a reproducible pipeline for ML experiments
summary
Shows how to convert a Jupyter Notebook prototype into a reproducible machine-learning pipeline using Git and DVC, illustrated with a text-to-image project built on Stable Diffusion. Covers modularizing notebook code into separate modules, composing those modules into a DVC pipeline, running configurable experiments through the pipeline, and comparing results to identify best outcomes.
AI-written summary · about
related talks
01h 32mHow to easily set up and version control your Machine Learning Pipelines
33mFlexible ML Experiment Tracking System for Python Coders with DVC and St...
39mNotebook To Production
22mDock Your Jupyter Notebook
53mTransforming your Jupyter Notebook to a Reproducible Machine Learning Pipeline using Kale
01h 18mProductionizing Jupyter Notebooks
59mDevOps for Machine Learning Projects
27mLeveraging Git for ML Experiment Management