-
Notifications
You must be signed in to change notification settings - Fork 3
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
BF: Add howto examples MuJoCo with image from camera #721
Conversation
@detlefarend @steveyuwono maybe useful for your OA stuff with CV. Run Howto bf_systems_004. This is also some preparation for my other stuff. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Hi Rizky, this is very interesting, inspiring, and relevant. It is a step ahead to open the black box MuJoCo and integrate parts. Could you please integrate it in rtd? If it is unsuitable for background processing/unit testing, you could refer to the src file directly in the howto.rst (like we did for howto_bf_streams_051_accessing_data_from_openml.py).
The howto raises the question of pooling and integrating CV components in MLPro. Currently, I think a CV component is more an adaptive function than a state-based system because it has no inner state in the simplest case. It maps an image to specific positional output information. That brings me to the point that we need a special StreamTask implementation that applies a (pre-trained) (adaptive) function to incoming stream data instances... A further StreamTask could integrate a state-based system to map an input instance (=action) to an output instance (=state)... However, this is a different story...
Done. |
Please also add a new cross reference entry to rtd-bf-systems-mujoco. It would also be nice to take a gif snapshot and add it to the howto description. The doc-string of the howto should also be taken over to the howto.rst. Thx! |
|
Description
Background
Checklists: