1 Repo
Converts discrete model tokens into continuous control commands for robotic actuators.
Distinct from Token Decoders: Specifically decodes tokens into robotic actuator commands rather than human-readable text
Explore 1 awesome GitHub repository matching artificial intelligence & ml · Action Command Decoding. Refine with filters or upvote what's useful.
OpenVLA is a vision-language-action model and framework designed for general-purpose robotic manipulation. It provides a robotic policy training framework and a control inference engine that map visual and textual inputs to robotic control actions, enabling zero-shot instruction following on hardware. The project includes a robotics dataset pipeline for standardizing diverse trajectory data and managing dataset mixtures. It supports large-scale model training through distributed GPU compute and sharded data parallelism, alongside parameter-efficient adaptation for fine-tuning models to new ta
Decodes discrete output tokens from the language-style model into continuous robotic control commands.