-
-
Notifications
You must be signed in to change notification settings - Fork 2.3k
Unable to install on Nvidia Xavier. Stuck at 100%. #4949
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
Comments
can you try
compilation takes time and I have a feeling it is taking time compiling things. @sciunto was it you taking care of the labels? I would like a label to collate installation issues so that we may better track challenges users have getting started on that front. |
also, your numpy version is very old. you definitely need to satisfy https://github.com/scikit-image/scikit-image/blob/master/requirements/build.txt I think below numpy 1.16, you will have a hard time on arm, but that is just my personal experience with that. |
What do you need specifically? |
with -vv it errors out with the following:
Guessing that scipy needs to be installed.
|
Install liblapack from your package manager. |
Did
from https://askubuntu.com/questions/300542/liblapack-so-file still getting
Also built from source sudo apt-get install gfortran Still getting the error. |
you likely need the dev files which are usually provided separately on your OS. Anyway, it doesn't seem to be an issue for scikit-image. |
I second @sciunto . You are compiling C code during the installation process which means you don't just need the shared libraries (obtained via At the very least you will need
and depending on other bindings you might need the header files for lapack and blas, too. ( |
For what it's worth.co da forge just aren't through this compilation procedure. Depending on your other dependencies, you might have some luck with conda and conda forge |
Here are the links to what I was talking about
Putting my conda-forge contributor hat on: while we (conda-forge) don't don't have the full list of packages for aarch64 compiled, we have quite a few and we recently got scikit-image. If your stack ends there, then it will really help boost you in your installation procedure. If you also need packages that we haven't compiled yet, for example opencv, then you will have to work on a more complete solution for your particular application. @FirefoxMetzger did give pointers in the right direction, So I think that will help. |
As we do not have further replies and the issue is not related to scikit-image, I'm closing. If not solved, feel free to reopen. |
Uh oh!
There was an error while loading. Please reload this page.
Description
Way to reproduce
on Nvidia Jetson Xavier
Version information
scikit-image installation stuck at 100%
When cloning and running pip3 install -e . Stuck at Obtaining file://foo/scikit-image
The text was updated successfully, but these errors were encountered: