Oier Mees
Oier Mees
Not sure if this is what you mean, but we provide an option to evaluate single tasks without resetting the robot to a neutral position. https://github.com/mees/calvin#multi-task-language-control-mtlc
Here you have an example of using task indicators to learn single task rl policies https://github.com/mees/calvin#reinforcement-learning-with-calvin
Feel free to send a PR if you have fixed this.
I am migrating my code to PL 2 and it seems that for the val dataloader getting a batch to be of the form {"key_a": batch_dataloader_a, "key_b": batch_dataloader_b} is not...
@ronghanghu could you comment on this issue please, would be very grateful for your help :)
you need to repace the argument 'refcoco' with 'coco_minus_refer', as this is the expected dataset string in the bash script
What issues do you have with the installation specifically?
> Hello, > > I am currently working on a project that involves the use of the CALIVIN benchmark. I am interested in performing secondary development on this dataset to...
Thanks! Could you provide the pretrained checkpoint in your repo so people can verify the reported numbers?
Besides, there are a few high performing related works that you are missing in your paper, would be nice to cite them in your paper too. 