Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Add pixel-wise options to grasp model, Integrate Spatial Softmax #295

Open
ahundt opened this issue Nov 4, 2017 · 1 comment
Open

Add pixel-wise options to grasp model, Integrate Spatial Softmax #295

ahundt opened this issue Nov 4, 2017 · 1 comment
Assignees
@ahundt
Labels
enhancement

Comments

@ahundt
Copy link
Member

ahundt commented Nov 4, 2017

Add pixel-wise training options to grasp_train and grasp_model, incorporating spatial softmax with the option to not find argmax during training, so that the image coordinate defined by the actual robot ending pose can be directly accessed for backpropagation.
@ahundt ahundt mentioned this issue Nov 4, 2017
Closed
10 tasks
@ahundt ahundt self-assigned this Nov 4, 2017
@ahundt
Copy link
Member Author

ahundt commented Jan 8, 2018 •
edited
Loading

There is some work on this in both tensorflow and keras-contrib:
tensorflow/tensorflow#12549

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees

@ahundt ahundt

Labels
enhancement
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
1 participant
@ahundt