The minimal conda cheatsheet
If you’re not using conda regularly, it’s almost certain you forget how to use it from usage to usage. I’ve compiled a list of commands I use most often.
You can download the PDF or Google doc version (fits on one page!).
Environment | Package |
Activate/switch | Install |
conda activate [env_name] conda activate pytorch_py36 |
conda install [package_spec ...] conda install numpy scikit-learn=0.23.1 |
Create | from specific channel |
conda create --name [env_name] [package_spec …] conda create --name math_py36 python=3.6 numpy |
conda install -c [channel] [package_spec …] conda install -c conda-forge numpy |
Create from file | Update package |
conda create --file [path] conda create --file exported_env.txt |
conda update [package_spec …] conda update numpy scikit-learn=0.23.1 |
Delete package | |
conda remove [package_name …] conda remove numpy scikit-learn |
|
* package_spec - package_name or package_name=package_version |
More about the same topic
-
Why requirements.txt isn't enough
-
Overview of python dependency management tools
-
Importing packages in Jupyter notebooks
-
The difference between conda and pip and how not to break your environment again?