Get vision+proprioceptive embeddings #38
Description
Hi, thanks for sharing your work. Could you please tell me how can I get the vision+proprioceptive embeddings from your model? I have a batch of samples where vision shape is batch size x time steps x 3 x 128 x128 and proprioceptive shape is batch size x time steps x features. I plan to use those embeddings for a downstream tasks. I assume embeddings from the semantically similar samples will be close to the each other in latent space.