V-JEPA: Learning Video Representations by Feature Prediction
Follow V-JEPA ViT-L/16 from video tubelets and feature prediction to pretrained encoder features, temporal tests and a small action-classification experiment.
The ExploringML journal
Notes, tutorials and experiments from my work with machine learning, AI and Python.
Subscribe via RSSBrowse the latest writing, or filter by a topic such as Python, MCP or deep learning.
Community
Share what you’re learning, ask questions, and swap ideas.
Join the ExploringML Discord.