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

Future Roadmap Suggestions #66

Open
RogerChern opened this issue Apr 5, 2019 · 13 comments
Open

Future Roadmap Suggestions #66

RogerChern opened this issue Apr 5, 2019 · 13 comments

Comments

@RogerChern
Copy link
Collaborator

@RogerChern RogerChern pinned this issue Apr 5, 2019
@xchani xchani unpinned this issue Apr 8, 2019
@xchani xchani pinned this issue Apr 8, 2019
@RogerChern
Copy link
Collaborator Author

RogerChern commented Apr 16, 2019 •
edited
Loading

@RogerChern RogerChern reopened this Apr 16, 2019
@xchani xchani unpinned this issue Apr 17, 2019
@xchani xchani pinned this issue Apr 17, 2019
@RogerChern
Copy link
Collaborator Author

RogerChern commented Apr 24, 2019 •
edited
Loading

  • TVM deployment

@RogerChern
Copy link
Collaborator Author

  • multi-node sync bn

@RogerChern
Copy link
Collaborator Author

  • horovod distributed training

@RogerChern
Copy link
Collaborator Author

RogerChern commented May 9, 2019 •
edited by xchani
Loading

@tusen-ai tusen-ai deleted a comment from feixiangdekaka May 15, 2019
@suman-19
Copy link

suman-19 commented Jul 6, 2019

Rotation Invariant Object Detection to train on DOTA dataset.

@RogerChern
Copy link
Collaborator Author

  • Support NDArray and Symbol dual modes for user-friendly debug.

@RogerChern
Copy link
Collaborator Author

@RogerChern
Copy link
Collaborator Author

RogerChern commented Aug 13, 2019 •
edited
Loading

@RogerChern
Copy link
Collaborator Author

@RogerChern
Copy link
Collaborator Author

  • wrap the contrib RoIAlign for our use
    Our one is written in quite an early day and is slightly different from the official Detectron implementation. Also, ours lacks FP16 support, sampling ratio, and PS variant.

@RogerChern
Copy link
Collaborator Author

pinning memory for loader

@xchani
Copy link
Contributor

xchani commented Jan 22, 2020 •
edited
Loading

  • model parallelism with pipeline

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

No branches or pull requests
3 participants
@RogerChern @xchani @suman-19