So this is possibly related to #1136 (pure speculation 馃槄 )
Basically, on a Vm with ubuntu 18:
conda create -n temp_env_scanpy
conda activate temp_env_scanpy
(temp_env_scanpy) giov@vm:~$ conda install -c bioconda scanpy
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.
Collecting package metadata (repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: -
Found conflicts! Looking for incompatible packages.
This can take several minutes. Press CTRL-C to abort.
failed
UnsatisfiableError: The following specifications were found to be incompatible with your CUDA driver:
- feature:/linux-64::__cuda==9.1=0
Your installed CUDA driver is: 9.1
Interestingly, this error is not thrown all the time, e.g. in a VM centos 7 without cuda:
Collecting package metadata (current_repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: failed with repodata from current_repodata.json, will retry with next repodata source.
Collecting package metadata (repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: \
Found conflicts! Looking for incompatible packages.
This can take several minutes. Press CTRL-C to abort.
failed
UnsatisfiableError:
Another student working with me had the same issue in windows. His error was:
UnsatisfiableError: The following specifications were found to be incompatible with your CUDA driver:
- feature:/win-64::__cuda==10.2=0
Your installed CUDA driver is: 10.2
But on a mac, no problem at all. In all situations, I have at least another environment with scanpy installed.
In all cases, conda was 4.8.3.
I cannot rule out completely the possibility that my conda in those 2 vms are messed up.
That's weird. Why would cuda be a dependency?
I'm not sure who is maintaining the bioconda recipe, so that might just be wrong. Do the new installation instructions work?
I had the issue trying to install it in a new environment when python itself wasn't installed there yet. After having installed python, it worked though. However, uing scanpy within a script doesn't work properly after the installation. I'll have a closer look at that now.
I am also unable to install scanpy on mac OS. I tried using python 3.8.x . 3.7.x and 3.6.x
(base) $ conda activate SCA
(SCA) $ conda --version
conda 4.8.2
(SCA) $ python --version
Python 3.6.10 :: Anaconda, Inc.
(SCA) $ conda install -c bioconda scanpy
Collecting package metadata (repodata.json): done
Solving environment: failed with initial frozen solve. Retrying with flexible solve.
Solving environment: \
Found conflicts! Looking for incompatible packages.
This can take several minutes. Press CTRL-C to abort.
failed
UnsatisfiableError: The following specifications were found to be incompatible with each other:
Output in format: Requested package -> Available versions
Make sure you are searching the conda-forge channel, too.
Either conda install -c conda-forge -c bioconda scanpy or configure the default channels.
Most helpful comment
Make sure you are searching the
conda-forgechannel, too.Either
conda install -c conda-forge -c bioconda scanpyor configure the default channels.