Install Intel® AI Analytics Toolkit via Conda*
Intel provides access to the AI Kit components through a public Anaconda repository. If you do not have an existing Conda-based python environment, install Conda or Miniconda*. To get more details on the AI Analytics Toolkit, visit the Intel AI Analytics toolkit home page.
Activate the conda base environment via opening Anaconda prompt or executing the following command in command prompt:
C:\<conda-install-dir>\anaconda3\Scripts\activate.bat
Create and activate your conda environment:
conda create -n <conda environment name> conda activate <conda environment name>
For Python 3.7, create an AI Kit Modin and machine learning environment named aikit-modin:
conda create -n aikit-modin -c intel intel-aikit-modin
For other Python versions, find the instructions at Intel® Distribution of Modin Getting Started Guide.
To install the packages from the Anaconda channel via Conda, execute the following command:
conda install <package_name>
For example, to install the TensorFlow* package, execute the following command:
conda install tensorflow
To install the packages from other channels, specify a channel name as in the example below:
conda install <package_name> -c <channel>
For example, to install the PyTorch* package from the pytorch channel, use the following command:
conda install pytorch -c pytorch
You can find the list of available packages and their channels in the section below.
List of Available Packages
Component Name |
Package Name |
Platform |
Channel |
|---|---|---|---|
Intel® Distribution for Python* |
intelpython3_full |
win |
intel |
Modin* |
modin-ray |
win |
anaconda |
Intel® Neural Compressor* |
neural-compressor (include -c conda forge on your command line) |
win |
intel, conda-forge |
PyTorch* |
pytorch |
win |
pytorch |
TensorFlow* |
tensorflow |
win |
anaconda |