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

About how to change model #7

Closed
MarkKang05 opened this issue Sep 13, 2019 · 4 comments
Closed

About how to change model #7

MarkKang05 opened this issue Sep 13, 2019 · 4 comments

Comments

@MarkKang05
Copy link

Hi! rwightman!
you're project is perfect.Its activated successfully on my laptop.
but my laptop is too slow.
so i try to change other model(more light, and more fast). I saw in your post that the weight is google's model is original. And i entered link (that you're linked, this -->> https://github.com/tensorflow/tfjs-models/tree/master/posenet )
but i couldnt found model.. and its just js code. I dont know how to convert to python code.
thank you..

@gzchenjiajun
Copy link

I also encountered this problem, I want to transfer the issue of resnet50 to python. Can you solve this problem?
@MarkKang05

@rwightman
Copy link
Owner

@MarkKang05 there is support for the different model multipliers of the mobilenet weights from the original tfjs release in the included scripts, just use a different --model argument

Available models via the argument (use the integer in the argument) include:

MOBILENET_V1_CHECKPOINTS = {
    50: 'mobilenet_v1_050',
    75: 'mobilenet_v1_075',
    100: 'mobilenet_v1_100',
    101: 'mobilenet_v1_101'
}

@gzchenjiajun
Copy link

I am watching people on the Internet saying that there is a version of resnet. How do I get the pre-training model of the resnet version? @rwightman

@rwightman
Copy link
Owner

@gzchenjiajun there is already an issue for that, #2 ... I have not ported the ResNet50 weights, the model format was changed so it requires more work and I haven't had the time

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

3 participants