Installing with conda#

Conda packages can be installed by running the following command:

conda install package> 

When conda installs a package, it is automatically added to your active environment. These packages are collections of files and directories that make up everything you need to use that particular library or software. For Python packages, these are primarily Python files that can be imported into other Python applications, but for compiled software packages, such as ffmpeg , these are typically binary executables you use directly on your computer.

If you would like to learn more about how environments are structured, head over to conda environments .

Below is a more precise overview of everything that happens during the installation process for a single package:

A graphic illustration of this process is shown below:

../../_images/installing-with-conda.png

Conda update versus conda install#

conda update updates packages to the latest compatible version. conda install can be used to install any version.

Conda uses the same rules for other packages. conda update always installs the highest version with the same major version number, whereas conda install always installs the highest version.

Installing conda packages offline#

To install conda packages offline, run: conda install /path-to-package/package-filename.tar.bz2/

If you prefer, you can create a /tar/ archive file containing many conda packages and install them all with one command: conda install /packages-path/packages-filename.tar

If an installed package does not work, it may be missing dependencies that need to be resolved manually.

Installing packages directly from the file does not resolve dependencies.

Installing conda packages with a specific build number#

If you want to install conda packages with the correct package specification, try pkg_name=version=build_string . Read more about build strings and package naming conventions. Learn more about package specifications and metadata.

For example, if you want to install llvmlite 0.31.0dev0 on Python 3.7.8, you would enter:

conda install -c numba/label/dev llvmlite=0.31.0dev0=py37_8