Installing Pruna with Pip and Conda
Prerequisites
pruna is currently only tested on Linux. We intend to support MacOS and Windows in the future. If you do not have a Linux machine on hand, you can still play around with pruna on e.g. Google Colab.
Conda is the recommended way to install pruna. You can use it to create and manage your python environments.
If you don’t already have it, you can download and install Miniconda from here. When you have Conda installed, you can create a new environment as follows:
conda create -n pruna python=3.10
conda activate pruna
If you are on a CUDA-enabled machine, you will need to make sure that the CUDA runtime toolkit is installed.
You can check that nvcc --version
provides a version that matches your nvidia-smi
version. If nvcc --version
fails or there is a version mismatch, you should install the correct CUDA runtime toolkit as follows:
conda install nvidia/label/cuda-12.1.0::cuda
or
conda install nvidia/label/cuda-11.8.0::cuda
Installing Pruna
If all prerequisites are met, you can simply install pruna with the following commands:
pip install pruna[gpu]==0.1.1 --extra-index-url https://prunaai.pythonanywhere.com/
pip install pruna[gpu-cu11]==0.1.1 --extra-index-url https://prunaai.pythonanywhere.com
pip install pruna==0.1.1 --extra-index-url https://prunaai.pythonanywhere.com --extra-index-url https://download.pytorch.org/whl/cpu
Additionally, if you intend to use methods for LLMs and ASR models, you should install the full version of pruna with the following command:
pip install pruna[full]==0.1.1 --extra-index-url https://prunaai.pythonanywhere.com
Now that you have installed pruna, get your pruna token and you are ready to go! You can then check out the Tutorials or craft your own example with our User Manual.