Skip to content
New issue

Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? # to your account

Can any deep learning model be used in nimble, such as yolov5?? #14

Open
umairjavaid opened this issue Aug 31, 2021 · 3 comments
Open

Comments

@umairjavaid
Copy link

umairjavaid commented Aug 31, 2021

No description provided.

@gyeongin
Copy link
Contributor

gyeongin commented Sep 1, 2021

I am not familiar with the YOLO v5 model, so let me list the general requirements of Nimble.
Nimble requires a neural network module to have the following properties:

  • CUDA only (should not entail any CPU operations)
  • Static (performs the same set of computations for different runs; e.g., the shape of the input image should be fixed)

@bgchun
Copy link
Collaborator

bgchun commented Sep 1, 2021

We can extend Nimble to relax the above requirements: handling CPU operations and dynamic shapes. However, these extensions are not part of the open-sourced Nimble.

@umairjavaid
Copy link
Author

Thank you so much!

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

No branches or pull requests

3 participants