Windows User Guide

Step 1: Install WSL2

For Windows, we recommend using Windows Subsystem for Linux 2 (WSL2) to run BigDL-Nano.

To install WSL2, you can now install everything you need by entering this command in an administrator PowerShell or Windows Command Prompt and then restarting your machine.

wsl --install

By default, this command should install the latest required components for WSL2 and install Ubuntu as default distribution for you.

To run this command, you must be running Windows 10 version 2004 and higher (Build 19041 and higher) or Windows 11. If you’re running an older build, or just prefer not to use the install command and would like step-by-step directions, see WSL manual installation steps for older versions.

To learn more about installation of WSL2, please Follow this guide.

Step 2: Install conda in WSL2

Start a new WSL2 window and set up the user information. Then download and install the conda.

wget https://repo.continuum.io/miniconda/Miniconda3-4.5.4-Linux-x86_64.sh
chmod +x Miniconda3-4.5.4-Linux-x86_64.sh
./Miniconda3-4.5.4-Linux-x86_64.sh

Step 3: Create a BigDL-Nano env

Use conda to create a new environment. For example, use bigdl-nano as the new environment name:

conda create -n bigdl-nano
conda activate bigdl-nano

Step 4: Install BigDL-Nano from Pypi

You can install BigDL-Nano from Pypi with pip. Specifically, for PyTorch extensions, please run:

pip install bigdl-nano[pytorch]
source bigdl-nano-init

For Tensorflow:

pip install bigdl-nano[tensorflow]
source bigdl-nano-init