Postprocessing and analysis of Cricksaw data
cellfinder install is described there: https://docs.brainglobe.info/cellfinder/installation/cluster-installation/slurm
It has been tested with cudatoolkit=11.2 cudnn=8.1
Following the first steps, you might get some library errors:
2021-12-19 20:40:30.766514: W tensorflow/stream_executor/platform/default/dso_loader.cc:64] Could not load dynamic library 'libcudart.so.11.0'; dlerror: libcudart.so.11.0: cannot open shared object file: No such file or directory; LD_LIBRARY_PATH: /usr/local/cuda/lib64
This can be fixed by pointing to the library folder of your cuda environment:
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:~/.conda/envs/cellfinder/lib/
Trying to update with pip --upgrade can break things.
When reinstalling I ran into ``GLIBCXX_3.4.30' not found ` errors. This can be fixed by installing the right version of gcc:
conda install -c conda-forge gccEasy:
pip install 'path/to/cricksaw/analysis'
Running elastix on camp requires a singularity container and elastix binaries
that are both in /camp/home/blota/home/shared/resources/elastix. See
scripts/analysis/cricksaw-register.sh for an example of how to use them.