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.

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

(feature) Geometry visualization of inputs/outputs #108

Closed
enricoros opened this issue May 12, 2018 · 2 comments
Closed

(feature) Geometry visualization of inputs/outputs #108

enricoros opened this issue May 12, 2018 · 2 comments

Comments

@enricoros
Copy link

Many papers show visually how an image input gets progressively shrunk but deepened by strided convolutions or expanded by deconvolutions.

Would be nice to have an isometric representation of the WHC input and output, in case of 3-dimensional geometries (batch=1).

The 2 isometric parallelepipeds could be drawn in the Node/Net panel.

@lutzroeder
Copy link
Owner

Yes, great idea. This would require #71 or is there another way to do this?

@enricoros
Copy link
Author

Yep, #71 is a milestone here. When I take a look at some TF models (mobilenets) I see Input and output dimensions correctly specified in the right panel - is it the case in which the model well specifies the tensor shapes?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants