Conda/Micromamba

The easiest way to export a .yml file from a conda environment is to first activate your environment by running:

conda activate myvenv

And then run the following command that generated your .yml file:

conda env export > list.yml

## win-64 and linux-64 have different label on version/number
## Export Conda Environment with minimized requirements
micromamba env export --from-history > list.yml

## like
name: uxarray
channels:
- anaconda
- conda-forge
dependencies:
- conda-forge::basemap
- conda-forge::cartopy
- anaconda::ipykernel
- conda-forge::netcdf4
- scipy
- conda-forge::uxarray==2024.11.1
- conda-forge::wrf-python

Finally, you can create a new environment using the .yml file, like:

conda env create --name myvenv --file=list.yml

or

conda env export | grep -v "^prefix: " > environment.yml
conda env create -f environment.yml

Remove environment,

conda remove -n ENV_NAME --all

Micromamba

Micromamba is a standalone version of Mamba which is an alternative to Conda. This stack delivers a custom anaconda/miniconda type deployment based on the communities conda-forge channel with micromamba as the package manager. For more information visit micromamba documentation.

Installation

wget micro.mamba.pm/install.sh
chmod +x install.sh
./install.sh

## Parsing arguments
if [ -t 0 ] ; then
  printf "Micromamba binary folder? [~/.local/bin] "
  read BIN_FOLDER
  printf "Init shell ($shell)? [Y/n] "
  read INIT_YES
  printf "Configure conda-forge? [Y/n] "
  read CONDA_FORGE_YES
fi

• /home/wpsze/.local/bin
• init shell? Y
• configure-forge? Y

Finally, it will initalize micromambs config on .bashrc

Export the contents of the existing environment to a YAML file.

micromamba env export -n oldenv > oldenv.yaml

Create env by given yaml

micromamba env create --name newenv --file oldenv.yaml

pip requirements file contain "@file"

• [Why does the pip requirements file contain @file instead of version number?](https://stackoverflow.com/questions/62586878/why-does-the-pip-requirements-file-contain-file-instead-of-version-number)
• https://peps.python.org/pep-0440/#direct-references

Source and activate env

$ source /home/wpsze/micromamba/etc/profile.d/micromamba.sh
$ micromamba activate venv

## For everyone to activate my env
$ micromamba activate /home/wpsze/micromamba/envs/mpich

activate or execute

Transaction finished

To activate this environment, use:

micromamba activate venv

Or to execute a single command in this environment, use:

micromamba run -n venv mycommand

clone an environment,

As stated in the answer to Cloning environment with micromamba, --clone is not an option in micromamba.

Instead, export the environment yaml and create a new environment using the --file flag. Then remove the old environment.

1. export the env dependencies: micromamba env export -n oldenv > oldenv.yaml
2. create a new env with a new name: micromamba create -n newenv --file oldenv.yaml
3. remove the old env: micromamba env remove -n oldenv
4. delete the yaml file: rm oldenv.yaml

check/add channel

$ micromamba info

# or

$ micromamba config list --sources

channels:
  - conda-forge  # '~/.condarc'

$ micromamba config append channels conda-forge

Cloning an environment

micromamba env create --name my_clone --clone new_venv

To access the remote machine with a browser

To access the remote machine with a browser the notebook must listen on an external facing port (not localhost). You will need the same invocation if want to run the Jupyter notebook on a container. In that case it is something like this:

jupyter-notebook --no-browser --port 11013 --ip=0.0.0.0

To listen only in localhost then you can omit the IP

jupyter notebook --no-browser --port=8080

References

1. 北京大学高性能计算校级公共平台用户文档