0

I am trying since yesterday to convert a grib file to a netcdf file (ECMWF dataset). As the grib file has to be interpolated from reduced to regular Gaussian grid before I can apply grib_to_netcdf, I will use CDO in my MacOS terminal. However, I cannot install the package in conda as I keep getting these messages:

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.

I have also tried to use jupyter notebook and python (with xarray) but since this file is roughly 66 GB, it is taking a long time and the kernel kept dying on me.

Has anyone experienced this before and can give me some advice? With pip install cdo it worked but then it's stored in the wrong channel (pypi instead of conda-forge) and if I try to call a cdo command in the terminal it says it's unknown.

Cheers!

Kim
  • 1
  • 1
  • I recommend switching to using mamba instead of conda. You will not run into these problems. Mambaforge is the best solution in my view https://github.com/conda-forge/miniforge – Robert Wilson Jul 03 '23 at 15:41
  • Note that the pypi Cdo is a python package which calls Cdo. It is not the same thing as the Cdo available on conda forge – Robert Wilson Jul 03 '23 at 15:42

0 Answers0