Skip to content

Improve data recording pipeline #45

Open
@rk1a

Description

@rk1a

The current data recorder is very basic. It simply dumps the raw data to disk.
It will make sense to store received observations in more structured and compressed formats, like video.
This is a prerequisite to creating datasets for imitation learning.

Metadata

Metadata

Assignees

Labels

No labels
No labels

Type

No type

Projects

Status

In Progress

Milestone

No milestone

Relationships

None yet

Development

No branches or pull requests

Issue actions