Conda-autoenv: Environments by Directory for Conda |
Automatically activate, update, and deactivate conda environments from an environment.yml file in a directory.
If a pip requirements.txt file exists as well, that will also be installed in the environment and updated upon deactivation.
Install ----
$ pip install conda-autoenv
$ echo "source `which conda_autoenv.sh`" >> ~/.bash_profile
Usage ----
Existing environment.yml file, with or without conda environment ~~~~~~~~~~~~~~~~~~~~~~~
If you already have an environment.yml file, place that in your directory (along with your pip requirements.txt file, if applicable), and conda-autoenv will automatically activate that environment and if applicable, create it beforehand.
No environment.yml file, with a conda environment ~~~~~~~~~~~~~~~~~~~~~~~
If you do not have an environment.yml file, but have a conda environment, export your conda environment to an environment.yml file by executing (in the environment's root directory, while your environment is activated):
$ conda env export > environment.yml
[Optional] You may also want your pip packages to automatically be activated and updated with your environment. If so, execute:
$ pip freeze > requirements.txt
No conda environment or environment.yml file (yet!) ~~~~~~~~~~~~~~~~~~~~~~~
If you do not have a conda environment, create one by executing (replace ENV_NAME with the name you would like to call your environment):
$ conda create -n ENV_NAME
Then, activate your environment by executing (replace ENV_NAME with the name you gave your environment):
$ source activate ENV_NAME
Finally, export your environment to an environment.yml file to be automatically activated later:
$ conda env export > environment.yml
[Optional] You may also want your pip packages to automatically be activated and updated with your environment. If so, execute:
$ pip freeze > requirements.txt