Skip to content

notes.md

Latest commit

 

History

History
128 lines (87 loc) · 2.46 KB

notes.md

File metadata and controls

128 lines (87 loc) · 2.46 KB

Anaconda

A faster version of conda, working with identical commands as described below, is mamba. We suggest to try this option.

To access a graphical interface to manage environment and launch utilities like jupyter:

anaconda-navigator

Conda

If during the anaconda installation conda has not been activated:

source $HOME/anaconda3/bin/activate

Update conda with:

conda update conda

Managing environments

Create a new environment with:

conda create --name env_name package_name
#example:
conda create --name snowflakes biopython

It can require to install additional packages, dependencies required by your packages.

Activate and deactivate

The environments can be activate and deactivate with:

conda activate env_name
conda activate #to go back to the base one
conda deactivate

To have the list of environments and paths:

conda info --envs

Different python versions

You can create environments with different versions of python:

conda create --name env_name python=2.7

A specific version of python can also be specified in a second moment, all the installed packages will be adapted to the required version of python (if available).

Managing packages

To look if a package is installed:

conda search package_name

To install it trough Anaconda repository:

conda install package_name
#pylab
conda install matplotlib
conda install scipy

If you want to see the list of installed packages:

conda list

Install custom packages

Move to the desired environment. Enter the local package folder, containing the setup.py file, then type:

pip install .

If some local package is modified you will simply need to reinstall the same package to override the previous one.

Delete an environment

conda env remove -n ENV_NAME
conda env remove --name ENV_NAME

Create an environment with a .yml file

The environment.yml file has to specify details of the packages, the environment name and the python version required.

name: envname
channels:
  - conda-forge
dependencies:
  - python>=3.9,<3.11
  - pyYAML
  - xarray
  - pip
  - pip:
    - sparse
    - git+https://github.com/jhardenberg/smmregrid.git

In this example a range of python versions is specified and some custom package that requires a download from a git repository has been added.

conda env create -f environment.yml