Skip to main content
  1. Posts/

Virtual Environment Management with Conda

··309 words·2 mins·
Table of Contents

Conda is the default package and environment management tool shipped with Anaconda.

This post logs my cheatsheet and notes for working with virtual environments with conda.

commonly used commands for virtual env
#

  • Create a new environment: conda create -n my_env
  • List environments on the system: conda env list
  • Delete an environment: conda env remove -n my_env
  • Activate an env:
    • Windows: conda activate my_env
    • Mac/Linux: source activate my_env
  • Exit/quit an env:
    • Windows: conda deactivate my_env
    • Mac/Linux: source deactivate my_env

How to install packages into a specific environment?
#

Outside an env, we can use the -n option to specify which environment we want to install packages into:

conda install -n my_env -c some_channel some_package

If we are inside the virtual environment, there is no need to use the -n option. We can install a package with conda install.

Remove package in an env
#

Remove pakcage from an env: conda remove -n my_env some_package. If you do not specify -n, some_package in current env will be removed. If you are not in any virtual env, some_package in the global Python path will be removed.

Using pip inside env install packages to global site-package location?
#

When I am inside a virtual env, I am surprised that pip install some_package installs the package to the global site-package instead of inside the virtual env.

After some digging, I find that reason is that I was not using the pip command inside that virtual environment. To check this, use the following command before using pip install:

which pip

If the pip path is not inside the virtual env, then we are using the system-wide pip. So pip install some_package will install some_package into to global Python path instead of the virtual environment.

References
#

Related

pip bad interpreter issue
··98 words·1 min
Configure Python logging with dictConfig
··503 words·3 mins
How to Profile Your Python Script/Module
·328 words·2 mins