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

Installation for Jetson Orin Nano on Jetpack 6.1 L4T36.4 #3133

Open
csvke opened this issue Oct 7, 2024 · 0 comments
Open

Installation for Jetson Orin Nano on Jetpack 6.1 L4T36.4 #3133

csvke opened this issue Oct 7, 2024 · 0 comments

Comments

@csvke
Copy link

csvke commented Oct 7, 2024

📚 The doc issue

Can someone please tell me how to install mmpose on a fresh Jetson Orin Nano?

The documentation didn’t specify the fact that pre-compiled mmcv are only for x86. So does it mean for Jetson, mmcv must be compiled from source?

As for compiling from source, I realise with 4GB RAM it’s very easy to run out of memory, so I have already got a 8GB swap setup.

And is it true that there’s no more mmcv-full from 2.x.x, only mmcv and mmcv-lite?

And I think if I use pip or mim to install to a Jetson, it will not find a .whl binary as no arm64 binaries are provided. So in that case, is pip / mim running the same process like ‘Compile mmcv from source’?

When compiling mmcv from source, do I only pass the argument ‘MMCV_WITH_OPS=1’ or are there more to make it CUDA enabled?

Does compiling mmpose depends on the version of mmcv?

Is there are more updated version of interoperability between
L4t / Python / PyTorch / CUDA against mmcv / mmdet / mmpose?

Does mmpose work with Jetpack 6 at all or do I have to downgrade to Jetpack 5?

Suggest a potential alternative/fix

No response

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

1 participant