Closed cgotelli closed 2 years ago
Hi @cgotelli - dependency resolution has definitely been an ongoing challenge with RivGraph, so your problem is understandable. I do not believe a specific conda version is needed at this time, however I'd check the version of Python you are trying to install RivGraph into. At this time we haven't explicitly tested Python 3.10, but 3.9 should be working, and I've just gotten it to install via the conda
instructions in a Google Colab notebook which is running Python 3.7.13.
To speed up the dependency resolution and installation processing, I would suggest trying mamba as an alternative to conda
. The installation instructions will be the same, you'd just replace the word conda
with mamba
in each line. In the Colab notebook, I was able to install RivGraph with the command !mamba install rivgraph -c conda-forge
in about 1 minute.
Hopefully that is helpful; if you still have trouble please provide a bit more context about your specific setup (Python version, operating system, etc.)
Thank you for your answer @elbeejay! I was able to install it in Colab, just like you did. So I tried to replicate the process, using also Python 3.7.13 in different pcs, environments, OS, etc. with no success. No clue about what is wrong, I tried all the possible alternatives I could find in internet to fix the endless Solving environment: failed with initial frozen solve.
but nothing worked. I was not even able to install mamba using conda-forge, and I had to install it with mamba-forge. Luckily, after doing that, it worked.
This situation is very strange. Not long ago I installed Rivgraph with no problems in the same pc, but after I had to format it, I could not make it work. I think the alternative you give is very useful and it would be good to have it in the documentation for future people like me.
Thanks again for your help. I will try to give Rivgraph a good use!
Closing. Thanks @elbeejay. One thing worth noting is that we migrated the rivgraph package to conda-forge (away from jschwenk) a couple of weeks ago, so make sure to pull from the conda-forge channel (as it's still available via jschwenk).
Hello! I have been trying to install with the different methods proposed in the website. However, it is forever processing without installing. Is it necessary a specific Conda version? I could not find any information about it. I also tried with a fresh installation of Anaconda and Miniconda, but it did not work ðŸ˜. Thanks in advance for your help.