Installation of Stardist and Cellpose in own conda env

IFB Core Cluster
,
1

Hello!
I am a new user of the FBI cluster and together with my team we are organizing a git and python bioimage analysis training on the cluster infrastructure.

I would like to create an exhausting conda environment that I can use during the course. I created one on my account and it worked fine, but for some reasons I don't understand I don't seem to be able to install Stardist or Cellpose, the process simply get killed but doesn't return any errors. These packages rely on PyTorch and Tensorflow so I assume that might be the issue, because if I try to install them directly the process is also killed. I also tried to clone an existing env but it also gets killed.

I tried to use Stardist and Cellpose from an already available conda env on the cluster, and it works fine, however I would like to be able to have everything in one place with a kernel connected to JupyterHub.

Could someone help me in this regard?
Thank you!

2

Hello Marco

cellpose is now installed by default in the Python 3.9 environment.
stardist is also installed by default in the Python 3.9 environment.

The Python 3.7 and 3.9 kernels are available from Jupyter

Note that when you do module load python/3.9 on the cluster, you have access to the same python as when you use the python 3.9 kernel from JupyterLab (the same python that integrates all the libraries).