arXiv - CS - Mathematical Software最新文献_第9页

A Sparse Fast Chebyshev Transform for High-Dimensional Approximation 高维逼近的稀疏快速切比雪夫变换

arXiv - CS - Mathematical Software Pub Date : 2023-09-26 DOI: arxiv-2309.14584

Dalton Jones, Pierre-David Letourneau, Matthew J. Morse, M. Harper Langston

引用次数: 0

pyPPG: A Python toolbox for comprehensive photoplethysmography signal analysis pyPPG:一个用于全面光容积脉搏波信号分析的Python工具箱

arXiv - CS - Mathematical Software Pub Date : 2023-09-24 DOI: arxiv-2309.13767

Marton A. Goda, Peter H. Charlton, Joachim A. Behar

{"title":"pyPPG: A Python toolbox for comprehensive photoplethysmography signal analysis","authors":"Marton A. Goda, Peter H. Charlton, Joachim A. Behar","doi":"arxiv-2309.13767","DOIUrl":"https://doi.org/arxiv-2309.13767","url":null,"abstract":"Photoplethysmography is a non-invasive optical technique that measures\u0000changes in blood volume within tissues. It is commonly and increasingly used\u0000for in a variety of research and clinical application to assess vascular\u0000dynamics and physiological parameters. Yet, contrary to heart rate variability\u0000measures, a field which has seen the development of stable standards and\u0000advanced toolboxes and software, no such standards and open tools exist for\u0000continuous photoplethysmogram (PPG) analysis. Consequently, the primary\u0000objective of this research was to identify, standardize, implement and validate\u0000key digital PPG biomarkers. This work describes the creation of a standard\u0000Python toolbox, denoted pyPPG, for long-term continuous PPG time series\u0000analysis recorded using a standard finger-based transmission pulse oximeter.\u0000The improved PPG peak detector had an F1-score of 88.19% for the\u0000state-of-the-art benchmark when evaluated on 2,054 adult polysomnography\u0000recordings totaling over 91 million reference beats. This algorithm\u0000outperformed the open-source original Matlab implementation by ~5% when\u0000benchmarked on a subset of 100 randomly selected MESA recordings. More than\u00003,000 fiducial points were manually annotated by two annotators in order to\u0000validate the fiducial points detector. The detector consistently demonstrated\u0000high performance, with a mean absolute error of less than 10 ms for all\u0000fiducial points. Based on these fiducial points, pyPPG engineers a set of 74\u0000PPG biomarkers. Studying the PPG time series variability using pyPPG can\u0000enhance our understanding of the manifestations and etiology of diseases. This\u0000toolbox can also be used for biomarker engineering in training data-driven\u0000models. pyPPG is available on physiozoo.org","PeriodicalId":501256,"journal":{"name":"arXiv - CS - Mathematical Software","volume":"10 6","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138521260","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Physics Informed Neural Network Code for 2D Transient Problems (PINN-2DT) Compatible with Google Colab 物理通知神经网络代码二维瞬态问题(PINN-2DT)与谷歌Colab兼容

arXiv - CS - Mathematical Software Pub Date : 2023-09-24 DOI: arxiv-2310.03755

Paweł Maczuga, Maciej Skoczeń, Przemysław Rożnawski, Filip Tłuszcz, Marcin Szubert, Marcin Łoś, Witold Dzwinel, Keshav Pingali, Maciej Paszyński

{"title":"Physics Informed Neural Network Code for 2D Transient Problems (PINN-2DT) Compatible with Google Colab","authors":"Paweł Maczuga, Maciej Skoczeń, Przemysław Rożnawski, Filip Tłuszcz, Marcin Szubert, Marcin Łoś, Witold Dzwinel, Keshav Pingali, Maciej Paszyński","doi":"arxiv-2310.03755","DOIUrl":"https://doi.org/arxiv-2310.03755","url":null,"abstract":"We present an open-source Physics Informed Neural Network environment for\u0000simulations of transient phenomena on two-dimensional rectangular domains, with\u0000the following features: (1) it is compatible with Google Colab which allows\u0000automatic execution on cloud environment; (2) it supports two dimensional\u0000time-dependent PDEs; (3) it provides simple interface for definition of the\u0000residual loss, boundary condition and initial loss, together with their\u0000weights; (4) it support Neumann and Dirichlet boundary conditions; (5) it\u0000allows for customizing the number of layers and neurons per layer, as well as\u0000for arbitrary activation function; (6) the learning rate and number of epochs\u0000are available as parameters; (7) it automatically differentiates PINN with\u0000respect to spatial and temporal variables; (8) it provides routines for\u0000plotting the convergence (with running average), initial conditions learnt, 2D\u0000and 3D snapshots from the simulation and movies (9) it includes a library of\u0000problems: (a) non-stationary heat transfer; (b) wave equation modeling a\u0000tsunami; (c) atmospheric simulations including thermal inversion; (d) tumor\u0000growth simulations.","PeriodicalId":501256,"journal":{"name":"arXiv - CS - Mathematical Software","volume":"16 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-09-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138521157","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Ensemble Differential Evolution with Simulation-Based Hybridization and Self-Adaptation for Inventory Management Under Uncertainty 不确定条件下基于仿真杂交的集成差分进化与自适应库存管理

arXiv - CS - Mathematical Software Pub Date : 2023-09-22 DOI: arxiv-2309.12852

Sarit Maitra, Vivek Mishra, Sukanya Kundu

{"title":"Ensemble Differential Evolution with Simulation-Based Hybridization and Self-Adaptation for Inventory Management Under Uncertainty","authors":"Sarit Maitra, Vivek Mishra, Sukanya Kundu","doi":"arxiv-2309.12852","DOIUrl":"https://doi.org/arxiv-2309.12852","url":null,"abstract":"This study proposes an Ensemble Differential Evolution with Simula-tion-Based\u0000Hybridization and Self-Adaptation (EDESH-SA) approach for inven-tory management\u0000(IM) under uncertainty. In this study, DE with multiple runs is combined with a\u0000simulation-based hybridization method that includes a self-adaptive mechanism\u0000that dynamically alters mutation and crossover rates based on the success or\u0000failure of each iteration. Due to its adaptability, the algorithm is able to\u0000handle the complexity and uncertainty present in IM. Utilizing Monte Carlo\u0000Simulation (MCS), the continuous review (CR) inventory strategy is ex-amined\u0000while accounting for stochasticity and various demand scenarios. This\u0000simulation-based approach enables a realistic assessment of the proposed\u0000algo-rithm's applicability in resolving the challenges faced by IM in practical\u0000settings. The empirical findings demonstrate the potential of the proposed\u0000method to im-prove the financial performance of IM and optimize large search\u0000spaces. The study makes use of performance testing with the Ackley function and\u0000Sensitivity Analysis with Perturbations to investigate how changes in variables\u0000affect the objective value. This analysis provides valuable insights into the\u0000behavior and robustness of the algorithm.","PeriodicalId":501256,"journal":{"name":"arXiv - CS - Mathematical Software","volume":"12 7","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138521256","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Unlocking massively parallel spectral proper orthogonal decompositions in the PySPOD package 在PySPOD包中解锁大规模平行光谱适当正交分解

arXiv - CS - Mathematical Software Pub Date : 2023-09-21 DOI: arxiv-2309.11808

Marcin Rogowski, Brandon C. Y. Yeung, Oliver T. Schmidt, Romit Maulik, Lisandro Dalcin, Matteo Parsani, Gianmarco Mengaldo

{"title":"Unlocking massively parallel spectral proper orthogonal decompositions in the PySPOD package","authors":"Marcin Rogowski, Brandon C. Y. Yeung, Oliver T. Schmidt, Romit Maulik, Lisandro Dalcin, Matteo Parsani, Gianmarco Mengaldo","doi":"arxiv-2309.11808","DOIUrl":"https://doi.org/arxiv-2309.11808","url":null,"abstract":"We propose a parallel (distributed) version of the spectral proper orthogonal\u0000decomposition (SPOD) technique. The parallel SPOD algorithm distributes the\u0000spatial dimension of the dataset preserving time. This approach is adopted to\u0000preserve the non-distributed fast Fourier transform of the data in time,\u0000thereby avoiding the associated bottlenecks. The parallel SPOD algorithm is\u0000implemented in the PySPOD (https://github.com/MathEXLab/PySPOD) library and\u0000makes use of the standard message passing interface (MPI) library, implemented\u0000in Python via mpi4py (https://mpi4py.readthedocs.io/en/stable/). An extensive\u0000performance evaluation of the parallel package is provided, including strong\u0000and weak scalability analyses. The open-source library allows the analysis of\u0000large datasets of interest across the scientific community. Here, we present\u0000applications in fluid dynamics and geophysics, that are extremely difficult (if\u0000not impossible) to achieve without a parallel algorithm. This work opens the\u0000path toward modal analyses of big quasi-stationary data, helping to uncover new\u0000unexplored spatio-temporal patterns.","PeriodicalId":501256,"journal":{"name":"arXiv - CS - Mathematical Software","volume":"15 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-09-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138521162","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Satisfiability.jl: Satisfiability Modulo Theories in Julia 可满足性。Julia中的可满足模理论

arXiv - CS - Mathematical Software Pub Date : 2023-09-15 DOI: arxiv-2309.08778

Emiko Soroka, Mykel J. Kochenderfer, Sanjay Lall

引用次数: 0

$texttt{ChisholmD.wl}$- Automated rational approximant for bi-variate series $texttt{ChisholmD.wl}$-双变量序列的自动有理逼近

arXiv - CS - Mathematical Software Pub Date : 2023-09-14 DOI: arxiv-2309.07687

Souvik Bera, Tanay Pathak

{"title":"$texttt{ChisholmD.wl}$- Automated rational approximant for bi-variate series","authors":"Souvik Bera, Tanay Pathak","doi":"arxiv-2309.07687","DOIUrl":"https://doi.org/arxiv-2309.07687","url":null,"abstract":"The Chisholm rational approximant is a natural generalization to two\u0000variables of the well-known single variable Pad'e approximant, and has the\u0000advantage of reducing to the latter when one of the variables is set equals to\u00000. We present, to our knowledge, the first automated Mathematica package to\u0000evaluate diagonal Chisholm approximants of two variable series. For the moment,\u0000the package can only be used to evaluate diagonal approximants i.e. the maximum\u0000powers of both the variables, in both the numerator and the denominator, is\u0000equal to some integer $M$. We further modify the original method so as to allow\u0000us to evaluate the approximants around some general point $(x,y)$ not\u0000necessarily $(0,0)$. Using the approximants around general point $(x,y)$,\u0000allows us to get a better estimate of the result when the point of evaluation\u0000is far from $(0,0)$. Several examples of the elementary functions have been\u0000studied which shows that the approximants can be useful for analytic\u0000continuation and convergence acceleration purposes. We continue our study using\u0000various examples of two variable hypergeometric series,\u0000$mathrm{Li}_{2,2}(x,y)$ etc that arise in particle physics and in the study of\u0000critical phenomena in condensed matter physics. The demonstration of the\u0000package is discussed in detail and the Mathematica package is provided as an\u0000ancillary file.","PeriodicalId":501256,"journal":{"name":"arXiv - CS - Mathematical Software","volume":"13 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-09-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138521244","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A Distributed Data-Parallel PyTorch Implementation of the Distributed Shampoo Optimizer for Training Neural Networks At-Scale 用于大规模训练神经网络的分布式洗发水优化器的分布式数据并行PyTorch实现

arXiv - CS - Mathematical Software Pub Date : 2023-09-12 DOI: arxiv-2309.06497

Hao-Jun Michael Shi, Tsung-Hsien Lee, Shintaro Iwasaki, Jose Gallego-Posada, Zhijing Li, Kaushik Rangadurai, Dheevatsa Mudigere, Michael Rabbat

{"title":"A Distributed Data-Parallel PyTorch Implementation of the Distributed Shampoo Optimizer for Training Neural Networks At-Scale","authors":"Hao-Jun Michael Shi, Tsung-Hsien Lee, Shintaro Iwasaki, Jose Gallego-Posada, Zhijing Li, Kaushik Rangadurai, Dheevatsa Mudigere, Michael Rabbat","doi":"arxiv-2309.06497","DOIUrl":"https://doi.org/arxiv-2309.06497","url":null,"abstract":"Shampoo is an online and stochastic optimization algorithm belonging to the\u0000AdaGrad family of methods for training neural networks. It constructs a\u0000block-diagonal preconditioner where each block consists of a coarse Kronecker\u0000product approximation to full-matrix AdaGrad for each parameter of the neural\u0000network. In this work, we provide a complete description of the algorithm as\u0000well as the performance optimizations that our implementation leverages to\u0000train deep networks at-scale in PyTorch. Our implementation enables fast\u0000multi-GPU distributed data-parallel training by distributing the memory and\u0000computation associated with blocks of each parameter via PyTorch's DTensor data\u0000structure and performing an AllGather primitive on the computed search\u0000directions at each iteration. This major performance enhancement enables us to\u0000achieve at most a 10% performance reduction in per-step wall-clock time\u0000compared against standard diagonal-scaling-based adaptive gradient methods. We\u0000validate our implementation by performing an ablation study on training\u0000ImageNet ResNet50, demonstrating Shampoo's superiority over standard training\u0000recipes with minimal hyperparameter tuning.","PeriodicalId":501256,"journal":{"name":"arXiv - CS - Mathematical Software","volume":"12 1","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138521248","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Integration of Quantum Accelerators with High Performance Computing $unicode{x2013}$ A Review of Quantum Programming Tools 量子加速器与高性能计算的集成$unicode{x2013}$量子编程工具综述

arXiv - CS - Mathematical Software Pub Date : 2023-09-12 DOI: arxiv-2309.06167

Amr Elsharkawy, Xiao-Ting Michelle To, Philipp Seitz, Yanbin Chen, Yannick Stade, Manuel Geiger, Qunsheng Huang, Xiaorang Guo, Muhammad Arslan Ansari, Christian B. Mendl, Dieter Kranzlmüller, Martin Schulz

{"title":"Integration of Quantum Accelerators with High Performance Computing $unicode{x2013}$ A Review of Quantum Programming Tools","authors":"Amr Elsharkawy, Xiao-Ting Michelle To, Philipp Seitz, Yanbin Chen, Yannick Stade, Manuel Geiger, Qunsheng Huang, Xiaorang Guo, Muhammad Arslan Ansari, Christian B. Mendl, Dieter Kranzlmüller, Martin Schulz","doi":"arxiv-2309.06167","DOIUrl":"https://doi.org/arxiv-2309.06167","url":null,"abstract":"Quantum computing (QC) introduces a novel mode of computation with the\u0000possibility of greater computational power that remains to be exploited\u0000$unicode{x2013}$ presenting exciting opportunities for high performance\u0000computing (HPC) applications. However, recent advancements in the field have\u0000made clear that QC does not supplant conventional HPC, but can rather be\u0000incorporated into current heterogeneous HPC infrastructures as an additional\u0000accelerator, thereby enabling the optimal utilization of both paradigms. The\u0000desire for such integration significantly affects the development of software\u0000for quantum computers, which in turn influences the necessary software\u0000infrastructure. To date, previous review papers have investigated various\u0000quantum programming tools (QPTs) (such as languages, libraries, frameworks) in\u0000their ability to program, compile, and execute quantum circuits. However, the\u0000integration effort with classical HPC frameworks or systems has not been\u0000addressed. This study aims to characterize existing QPTs from an HPC\u0000perspective, investigating if existing QPTs have the potential to be\u0000efficiently integrated with classical computing models and determining where\u0000work is still required. This work structures a set of criteria into an analysis\u0000blueprint that enables HPC scientists to assess whether a QPT is suitable for\u0000the quantum-accelerated classical application at hand.","PeriodicalId":501256,"journal":{"name":"arXiv - CS - Mathematical Software","volume":"18 2","pages":""},"PeriodicalIF":0.0,"publicationDate":"2023-09-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"138521079","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

CDL: A fast and flexible library for the study of permutation sets with structural restrictions CDL:一个快速灵活的库，用于研究具有结构限制的置换集

arXiv - CS - Mathematical Software Pub Date : 2023-09-12 DOI: arxiv-2309.06306

Bei Zhou, Klas Markstrōm, Søren Riis

引用次数: 0