Introducing Decision Transformers on Hugging Face 🤗
Hugging Face Blog
Hugging Face integrated Decision Transformer, an offline reinforcement learning method, into its transformers library and Hub. The library now includes nine pre-trained model checkpoints for continuous control tasks in Gym environments such as Hopper, Walker2D, and Halfcheetah. Users can now load and deploy Decision Transformer models for offline RL applications where agents learn from existing datasets rather than through real-time environment interaction.