Subpackages#

Module contents#

robomimic.register_all_links()#

Record all dataset links in this function.

robomimic.register_all_momart_links()#

Record all dataset links in this function.

robomimic.register_dataset_link(task, dataset_type, hdf5_type, link, horizon)#

Helper function to register dataset link in global dictionary. Also takes a @horizon parameter - this corresponds to the evaluation rollout horizon that should be used during training.

Parameters

• task (str) – name of task for this dataset

• dataset_type (str) – type of dataset (usually identifies the dataset source)

• hdf5_type (str) – type of hdf5 - usually one of “raw”, “low_dim”, or “image”, to identify the kind of observations in the dataset

• link (str) – download link for the dataset

• horizon (int) – evaluation rollout horizon that should be used with this dataset

robomimic.register_momart_dataset_link(task, dataset_type, link, dataset_size)#

Helper function to register dataset link in global dictionary. Also takes a @horizon parameter - this corresponds to the evaluation rollout horizon that should be used during training.

Parameters

• task (str) – name of task for this dataset

• dataset_type (str) – type of dataset (usually identifies the dataset source)

• link (str) – download link for the dataset

• dataset_size (float) – size of the dataset, in GB