Work

PyTorchSim: A Comprehensive, Fast, and Accurate NPU Simulation Framework

Year: 2025

Type: conference-paper

Abstract: This is the Docker image for the artifact of "PyTorchSim: A Comprehensive, Fast, and Accurate NPU Simulation Framework", published in MICRO'25. You can get the latest code and docker image from our Github. Note: Instead of using Zenodo’s Download all, downloading split files individually in parallel was significantly faster in our experience. After downloading all parts, the image can be loaded into Docker using the following command: $ cat torchsim_image.part_* | docker load $ docker run -it --ipc=host -w /workspace/PyTorchSim ghcr.io/psal-postech/torchsim-ci:v1.0.0 bash

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.16419767

Authors: Yang, Wonhyuk, Shin, Yunseon, Woo, Okkyun, Park, Geonwoo, Ham, Hyungkyu, Kang, Jeehoon, Park, Jongse, Kim, Gwangsun

Institutions: Pohang University of Science and Technology, Samsung Electronics (South Korea), Korea Advanced Institute of Science and Technology

Language: English

Topic: Parallel Computing and Optimization Techniques

Subfield: Hardware and Architecture

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W6949692327

FWCI:

Cites:

Cited by:

Related to: