Installation¶
Installing from package managers¶
PyPi¶
As seen in the previous quickstart examples, we can install FedRAG via pip
:
Conda¶
For conda
users, fed-rag
has been published to the
conda-forge
channel, and thus can be installed
with conda
using the below command:
Installing from source¶
To install from source, first clone the repository:
# https
git clone https://github.com/VectorInstitute/fed-rag.git
# ssh
git clone git@github.com:VectorInstitute/fed-rag.git
After cloning the repository, you have a few options for installing the library.
The next two subsections outline how to complete the installation using either
pip
or uv
, respectively.
Using pip
¶
To complete the installation, first cd
into the fed-rag
directory and then
run the following pip install
command:
Tip
We recommended to always use a fresh virtual environment for new projects. Before running the above command, ensure that your dedicated virtual environment is active.
Using uv
¶
FedRAG uses uv
for dependency management, publishing
to PyPi, and for setting up development environments.
Users can also use uv
to complete the source installation of FedRAG.
Note
This method requires uv
to be installed onto the users development machine.
For installation instructions visit uv
's official documentation.
To install with desired extras and groups, add the flags --extra <extra-name>
and --optional <optional-name>
, respectively. As an example: