NA-MIC Project Weeks
Back to Projects List
Enabling PyTorch3D on Windows and Optimizing Miniconda for Slicer Extensions
Key Investigators
- Leroux Gaelle (University of Michigan, USA)
- Claret Jeanne (University of Michigan, USA)
- Cevidanes Lucia (University of Michigan, USA)
- Hutin Nathan (CPE Lyon, France)
- Allemand David (Kitware, USA)
- Prieto Juan Carlos (University of North Carolina, USA)
Project Description
This project focuses on enhancing compatibility and usability in two key areas. Firstly, we aim to enable the use of PyTorch3D on the Windows platform. By leveraging the Windows Subsystem for Linux (WSL2) and a virtual Miniconda environment, we intend to bypass the traditional limitations and provide Windows users with full access to PyTorch3D’s capabilities. Secondly, the project seeks to improve the integration of Miniconda with Slicer extensions. Our goal is to simplify the process of creating and managing virtual environments for Slicer extensions, thereby making the procedure more intuitive. This will not only ease the use of various analytical tools and libraries but also streamline the setup process within WSL, especially for tools incompatible with Windows. This approach aims to bridge the gap in functionality and user experience across different platforms.
Objective
Our project aims to achieve two primary objectives:
Operationalizing PyTorch3D on Windows: The first goal is to make PyTorch3D, typically unsupported on the Windows platform, fully functional. We plan to employ the Windows Subsystem for Linux (WSL2) combined with a virtual Miniconda environment to overcome this limitation. This strategy is designed to provide Windows users with complete access to the extensive functionalities of PyTorch3D.
Improving Miniconda Integration for Slicer Extensions: Our second objective is to enhance the use of Miniconda as a virtual environment manager specifically for Slicer extensions. We aim to streamline the process of creating and managing new virtual environments that are utilized by Slicer extensions, making the procedure more intuitive and user-friendly. This advancement will facilitate the use of specialized libraries required for a variety of analytical tools, which are currently not integrable directly into Slicer. Additionally, this approach will assist in the setup of Miniconda3 and the creation of new environments within WSL, particularly for tools that are not available on Windows.
Approach and Plan
- Investigate PyTorch3D and Windows compatibility, pinpointing causes of incompatibility.
- Catalog PyTorch3D dependencies for WSL2 and outline requisite system configurations.
- Operationalize PyTorch3D on Windows using WSL2 and Miniconda.
- Develop an Automated Installer for WSL2 Setup on Windows.
- Create a Module to Streamline Miniconda for Slicer Extensions
Progress and Next Steps
- Completed Initial Research on PyTorch3D and WSL2 Compatibility
- Successfully ran PyTorch3D on WSL2
- Created an installer for WSL2 Next Step :
- Development of a new Slicer module for managing Miniconda
Illustrations
No response
Background and References
No response