google-deepmind / google-deepmind/open_x_embodiment
Testing on new dataset
Open
- Dominant language
- Jupyter Notebook
- Stars
- 2k
- Forks
- 127
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
Can you provide instructions on how to create a new dataset to use with the pretrained model?
According to the paper, the model can support either absolute and relative positions as well as velocities.
How is this implemented/supported by the model/dataset?
Do I need to compensate for the different viewpoints in each dataset in any way?
Thanks,
Contributor guide
Assessment
This issue has not been assessed yet.