Start Locally

Select your preferences and run the install command. Stable represents the most currently tested and supported version of pysensing. This should be suitable for many users. Please ensure that you have met the prerequisites below (e.g., numpy), depending on your package manager.

pysensing Build
Your OS
Package
Language
Compute Platform
Run this Command:
pysensing Build
Stable (0.2.0)
Your OS
Linux
Windows
Pip
Language
Python
Compute Platform
CUDA 10.2
CUDA 11.3
Run this Command:
pip install pysensing

Installing on macOS

pysensing can be installed and used on macOS. Depending on your system and compute requirements, your experience with pysensing on a Mac may vary in terms of processing time. It is recommended, but not required, that your Mac have an NVIDIA GPU in order to harness the full power of pysensing’s CUDA support.

Currently, CUDA support on macOS is only available by building pysensing from source

Prerequisites

macOS Version

pysensing is supported on macOS 10.10 (Yosemite) or above.

Python

It is recommended that you use Python 3.7 or greater, which can be installed either through the Anaconda package manager (see below), Homebrew, or the Python website.

Package Manager

To install the pysensing binaries, you will need to use one of two supported package managers: Anaconda or pip. Anaconda is the recommended package manager as it will provide you all of the pysensing dependencies in one, sandboxed install, including Python.

Anaconda

To install Anaconda, you can download graphical installer or use the command-line installer. If you use the command-line installer, you can right-click on the installer link, select Copy Link Address, and then use the following commands:

# The version of Anaconda may be different depending on when you are installing`
curl -O https://repo.anaconda.com/miniconda/Miniconda3-latest-MacOSX-x86_64.sh
sh Miniconda3-latest-MacOSX-x86_64.sh
# and follow the prompts. The defaults are generally good.`

pip

Python 3

If you installed Python via Homebrew or the Python website, pip was installed with it. If you installed Python 3.x, then you will be using the command pip3.

Tip: If you want to use just the command pip, instead of pip3, you can symlink pip to the pip3 binary.

Installation

pip

To install pysensing via pip, use one of the following command

# Python 3.x
pip install pysensing

Verification

To ensure that pysensing was installed correctly, we can verify the installation by running sample pysensing code. Here we will construct a randomly initialized tensor.

>>> import torch, pysensing as pp

>>> # A random so(3) LieTensor
>>> r = pp.randn_so3(2, requires_grad=True)
    so3Type LieTensor:
    tensor([[ 0.1606,  0.0232, -1.5516],
            [-0.0807, -0.7184, -0.1102]], requires_grad=True)

Building from source

For the majority of pysensing users, installing from a pre-built binary via a package manager will provide the best experience. However, there are times when you may want to install the bleeding edge pysensing code, whether for testing or actual development on the pysensing core.

You will also need to build from source if you want CUDA support.

Prerequisites

  1. Install Anaconda
  2. Install CUDA, if your machine has a CUDA-enabled GPU.
  3. Install from source.

You can verify the installation as described above.

Installing on Linux

pysensing can be installed and used on various Linux distributions. Depending on your system and compute requirements, your experience with pysensing on Linux may vary in terms of processing time. It is recommended, but not required, that your Linux system has an NVIDIA GPU in order to harness the full power of pysensing's CUDA support.

Prerequisites

Python

Python 3.7 or greater is generally installed by default on any of our supported Linux distributions, which meets our requirements. We recommend starting with Anaconda and pip to create a controlled environment for your pysensing installation.

Install Anaconda

Download and install Anaconda using the command-line installer:


# The version of Anaconda may be different depending on when you are installing
curl -O https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
sh Miniconda3-latest-Linux-x86_64.sh

Follow the prompts. The defaults are generally good. You may need to open a new terminal or re-source your ~/.bashrc to get access to the conda command.

Create a pysensing environment with the specified Python version. We recommend using Python >= 3.10:

conda create -n pysensing python=3.10
conda activate pysensing

Install PyTorch

Follow the official PyTorch installation guide to install PyTorch based on your system's capabilities:

For systems with CUDA support (replace cu118 with your CUDA version):

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu118

For systems without CUDA support:

pip install torch torchvision torchaudio

Installation

pip

The pysensing package itself can be installed from PyPI:

pip install pysensing

Building from Source

If you want to build from source and get the latest version:

pip install --upgrade git+https://github.com/pysensing/pysensing.git

Verification

To ensure that pysensing was installed correctly, verify the installation by running a sample tutorial notebook:

  1. Clone the pysensing repository:
    2. git clone https://github.com/pysensing/pysensing.git
  2. Install Jupyter via pip:
    3. pip install jupyter
  3. Navigate to the tutorial directory and launch Jupyter Notebook:
    4. cd pysensing/pysensing/acoustic/tutorials/
jupyter notebook

This will open the tutorial notebooks in your default web browser.

Installing on Windows

pysensing can be installed and used on various Windows distributions. Depending on your system and compute requirements, your experience with pysensing on Windows may vary in terms of processing time. It is recommended, but not required, that your Windows system has an NVIDIA GPU in order to harness the full power of pysensing's CUDA support.

Prerequisites

Supported Windows Distributions

pysensing is supported on the following Windows distributions:

The install instructions here will generally apply to all supported Windows distributions. The specific examples shown will be run on a Windows 11 Home machine.

Python

Currently, pysensing on Windows only supports Python 3.7 or greater; Python 2.x is not supported.

As it is not installed by default on Windows, there are multiple ways to install Python:

If you decide to use Chocolatey and haven't installed Chocolatey yet, ensure that you are running your command prompt as an administrator.

For a Chocolatey-based install, run the following command in an administrative command prompt:

# Install Python via Chocolatey
choco install python

Package Manager

To install the pysensing binaries, you will need to use at least one of two supported package managers: Anaconda and pip.

pip

If you installed Python by any of the recommended ways above, pip will have already been installed for you.

Anaconda

Follow the instructions in Anaconda Download, and you will install Anaconda, including the Anaconda Navigator, which is intuitive for package management and recommended for Windows users.

Make sure you have added your conda-directory and conda-directory/Scripts to your PATH variables after Anaconda installation. Otherwise, the conda command in the Windows command-line tool will not be recognized.

CUDA

Before installing PyTorch, it is better to check the compatibility of your device. Refer to the CUDA compatibility and NVIDIA support matrix.

Note that CUDA Toolkit should be installed first, as it is the foundation for cuDNN.

CUDA Toolkit

Follow the instructions in CUDA Toolkit and run the executable for installation.

After installation, ensure the environment variables CUDA_PATH and CUDA_PATH_V12_1 (or your CUDA version) are set in the Windows system variables. Check the success of the installation with nvcc -V.

cuDNN

When the CUDA Toolkit is ready, download cuDNN, extract the contents, and copy the lib, bin, and include directories into your CUDA installation directory. Update your PATH variables accordingly.

PyTorch

Create a virtual environment for PyTorch and pysensing installation:

# Create and activate a conda environment
conda create -n your_dev python=3.10 
conda activate your_dev

Install PyTorch:

For systems with CUDA support:

pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu121

For systems without CUDA support:

pip install torch torchvision torchaudio

Installation

pip

The pysensing package itself can be installed from PyPI:

pip install pysensing

Building from Source

If you want to build from source and get the latest version:

pip install --upgrade git+https://github.com/pysensing/pysensing.git

Verification

To ensure that pysensing was installed correctly, verify the installation by running a sample tutorial notebook:

  1. Clone the pysensing repository:
    2. git clone https://github.com/pysensing/pysensing.git
  2. Install Jupyter via pip:
    3. pip install jupyter
  3. Navigate to the tutorial directory and launch Jupyter Notebook:
    4. cd pysensing/pysensing/acoustic/tutorials/
jupyter notebook

This will open the tutorial notebooks in your default web browser.