Closed lxsteiner closed 1 year ago
Hi, the conda installation doesn't work currently due to dependency issues with mkl and pytorch. The best way to install is from source or using docker (see https://ggcaller.readthedocs.io/en/latest/installation.html). The depreciation warnings should be fine, please let me know if you run into any issues!
Closing as no activity.
Hi,
tried to install with conda recipe from bioconda:
mamba create -n ggcaller_env -c defaults -c bioconda -c conda-forge ggcaller
after installation it fails:Here are the package versions inside the environment:
System info:
Saw in the other closed issues that the biopython version caused issues in some cases?
Then tried installing from source via the .yml file and the github repo. After
$ python setup.py install
there were some warnings, but seems to have installed OK so far (at least the help message is printing):Plenty of these
[-Wdeprecated-declarations]
warnings... I can provide you with the entire log if interested or relevant; I'm not familiar with the warnings.Here the package versions in this (hopefully functioning?) environment, for comparison:
Biopython version is certainly different (1.78 vs 1.80). Not sure if that was the only cause of the previous issues, python is also different (3.10 vs 3.9).
I'll try to do a test run with the dataset from the tutorial and let you know if I run into further issues. Thank you!