PyCaret 2.1 is now available. Click here to see release notes. Documentation on the website is only updated for major releases. To see the latest documentation, Click Here

Installing the latest release

Installing PyCaret is the first step towards building your first machine learning model in PyCaret. Installation is easy and takes only a few minutes. All dependencies are also installed with PyCaret. Click here to see the complete list of dependencies. 

In order to avoid potential conflicts with other packages it is strongly recommended to use a virtual environment, e.g. python3 virtualenv (see python3 virtualenv documentation) or conda environments. Using an isolated environment makes it possible to install a specific version of pycaret and its dependencies independently of any previously installed Python packages. See an example below of how to create a conda environment and install PyCaret. 

#create a conda environment
conda create --name yourenvname python=3.6

#activate environment
conda activate yourenvname

#install pycaret
pip install pycaret

#create notebook kernel connected with the conda environment
python -m ipykernel install --user --name yourenvname --display-name "display-name"


The following libraries have been removed from hard dependency in PyCaret 2.0. Hence they must be installed separately when specific functionalities are being used. See the code below on how to install these dependencies.

# install shap for interpret_model functionality 
pip install shap 

# if build for shap fails using pip: 
conda install -c conda-forge shap 

# install awscli for deploy_model functionality 
pip install awscli 

# install azure-storage-blob for deploy_model
pip install azure-storage-blob

# install google-cloud-storagefor deploy_model
pip install google-cloud-storage

# install psutil for system logging 
pip install psutil


PyCaret Nightly Build

PyCaret is a fast-evolving machine learning library. Often, you want to have access to the latest features but want to avoid compiling PyCaret from source or waiting for the next release. Fortunately, you can now install pycaret-nightly using pip.

We highly recommend to install pycaret-nightly in a virtual environment to avoid conflicts.

# create a conda environment
conda create --name yourenvname python=3.6

# activate environment
conda activate yourenvname

# install pycaret
pip install pycaret-nightly

# update pycaret-nightly
pip install --upgrade pycaret-nightly

# create notebook kernel connected with the conda environment
python -m ipykernel install --user --name yourenvname --display-name "display-name"


Recommended Environment for use

You can use PyCaret in your choice of Integrated Development Environment (IDE) but since it uses html and several other interactive widgets, it is optimized for use within notebook environment, be it Jupyter NotebookJupyter LabAzure Notebooks or Google Colab.

Learn how to install Jupyter Notebook
Learn how to install Jupyter Lab
Get Started with Azure Notebooks
Get Started with Google Colab
Get Started with Anaconda Distribution


Run a PyCaret Container

A Docker container runs in a virtual environment and is the easiest way deploy applications using PyCaret. Dockerfile from base image python:3.7 and python:3.7-slim is tested for PyCaret 2.0.

FROM python:3.7-slim


ADD . /app

RUN apt-get update && apt-get install -y libgomp1

RUN pip install --trusted-host -r requirements.txt

CMD pytest #replace it with your entry point.


Building from source

You can also download the source file from the link below and use the pip installer to install the package from a downloaded file. To install the package using the source file, download the file and use the command line or notebook environment to run the below cell of code.

# install using source file
pip install C:/path_to_download/pycaret-version.tar.gz

# install directly from git (stable release)
pip install

For Google Colab

PyCaret uses interactive plotting ability. In order to render interactive plots in Google Colab, run the below line of code in your colab notebook.

#For Google Colab only
from pycaret.utils import enable_colab 



MAC users will have to install LightGBM separately using Homebrew, or can be built using CMake and Apple Clang or gcc. See the instructions below:

  1. Install CMake (3.16 or higher)
    >> brew install cmake
  2. Install OpenMP
  3. >> brew install libomp
  4. Run the following command in terminal:
git clone --recursive ; cd LightGBM
mkdir build ; cd build
cmake ..
make -j4


Try this next