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

How to get your weights? #8

Open
JohnCraigPublic opened this issue May 11, 2020 · 3 comments
Open

How to get your weights? #8

JohnCraigPublic opened this issue May 11, 2020 · 3 comments

Comments

@JohnCraigPublic
Copy link

Hello - would like to try our your demo! Where can I find the model weights that you load?

@JohnCraigPublic
Copy link
Author

Also, can't find tensorflow_addons -- where does that come from? Thanks!!

@JohnCraigPublic
Copy link
Author

You say:
"Once you have all the images and annotations, copy tfrecords_mpii.py to the new dataset folder and run it to generate TF Records"
But I don't find tfrecords_mpii.py help! Need a hint from you... Thanks...

@ethanyanjiali
Copy link
Owner

Hey John, tfrecords_mpii.py is here: https://github.com/ethanyanjiali/deep-vision/tree/master/Datasets/MPII

Unlike COCO or ImageNet, MPII is a relatively small dataset, so you should be able to train it quickly. My implementation in this repo is mostly for study purpose, so it doesn't yield similar PCKh as in the paper. I would recommend this repo if you are looking for pretrained weights: https://github.com/princeton-vl/pytorch_stacked_hourglass
I also have a private implementation with comparable metrics, and the main difference is the data augmentation part.

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