Hacker News with Generative AI: Scientific Computing

Neuralatex: A machine learning library written in pure LATEX (neuralatex.com)
Neuralatex is a scalar values-based auto-grad library similar to MicroGrad but written entirely in latex! As part of your latex document you can specify the architecture of a neural network and loss functions, how to generate or load training data, and specify training hyperparameters and experiments. When the document is compiled, the latex compiler will generate or load training data, train the network, run experiments and generate figures.
Frink (frinklang.org)
Frink is a practical calculating tool and programming language designed to make physical calculations simple, to help ensure that answers come out right, and to make a tool that's really useful in the real world.
Datoviz: High-Performance GPU Scientific Visualization Library with Vulkan (khronos.org)
Datoviz is a cross-platform, open-source, high-performance GPU scientific data visualization library designed for interactive exploration of large datasets.
Fastplotlib: GPU-accelerated, fast, and interactive plotting library (medium.com)
fastplotlib is a new GPU-accelerated fast and interactive scientific plotting library that leverages WGPU
New horizons for Julia (lwn.net)
Julia, a free, general-purpose programming language aimed at science, engineering, and related arenas of technical computing, has steadily improved and widened its scope of application since its initial public release in 2012.
Understanding Automatic Differentiation in Jax: A Deep Dive (ispeakcode.substack.com)
Welcome to the world of JAX, where differentiation happens automatically, faster than a caffeine-fuelled coder at 3 a.m.! In this post, we’re going to delve into the concept of Automatic Differentiation (AD), a feature at the heart of JAX, and we’ll explore why it’s such a game changer for machine learning, scientific computing, and any other context where derivatives matter. The popularity of JAX has been increasing lately, thanks to the emerging field of scientific machine learning powered by differentiable programming.
Bloodflowtrixi.jl – 1D and 2D blood flow models for arterial circulation (github.com/yolhan83)
BloodFlowTrixi.jl is a Julia package that implements one-dimensional (1D) and two-dimensional (2D) blood flow models for arterial circulation.
The Tensor Cookbook (2024) (tensorcookbook.com)
This book aims to standardize the notation for tensor diagrams by rewriting the classical "Matrix Cookbook" using this notation.
Etalumis: Bringing Probabilistic Programming to Scientific Simulators at Scale (arxiv.org)
Probabilistic programming languages (PPLs) are receiving widespread attention for performing Bayesian inference in complex generative models.
Scientific Computing on the Sony Playstation 2 (2003) (archive.org)
The National Center for Supercomputing Applications and the Computer Science department at the University of Illinois are exploring the use of the Sony PlayStation® 2 game console for scientific computing and high-resolution visualization.
Scientific computing with confidence using typed dimensions (laurentrdc.xyz)
I have performed non-trivial scientific calculations, in university and beyond, for almost 15 years.
Khronos SYCL Being Updated to Increase Appeal for HPC and Scientific Computing (phoronix.com)
In addition to the release today of OpenMP 6.0 ahead of the SC24 supercomputing conference in Atlanta, over at The Khronos Group they have provided an update on upcoming SYCL improvements to benefit high performance computing (HPC) and scientific computing applications.
Ask HN: Are my HPC professors right? Is Python worthless compared to C? (ycombinator.com)
I'm a PhD student implementing a finite element code. It simulates electromagnet waves passing through heterogeneous material. This code has to run in parallel, and run fast. I've been using old C libraries like PETSc to do this, and honestly, I do not enjoy working with C at all. Its esoteric and difficult to understand, and just overall feels like I'm using a tool from the 70s.
My NumPy year: Creating a DType for the next generation of scientific computing (quansight.com)
From no CPython C API experience to shipping a new DType in NumPy 2.0.
My NumPy year: Creating a DType for the next generation of scientific computing (quansight.com)
From no CPython C API experience to shipping a new DType in NumPy 2.0.
Machine Learning to Computational Plasma Physics Reduced-Order Plasma Modeling (arxiv.org)
Machine learning (ML) provides a broad spectrum of tools and architectures that enable the transformation of data from simulations and experiments into useful and explainable science, thereby augmenting domain knowledge.
Building a compile-time SIMD optimized smoothing filter (scientificcomputing.rs)
I built a Savitzky-Golay filter (fancy name for a dot product with some known constants on a rolling window) and tried to optimize the crap out of it.
CuPy: NumPy and SciPy for GPU (github.com/cupy)
CuPy is a NumPy/SciPy-compatible array library for GPU-accelerated computing with Python.
Performance of Eigen vs. Blaze vs. Fastor vs. Armadillo vs. XTensor (2020) (medium.com)
It is March 2020. C++20 is almost around the corner and as a scientific C++ programmer I am quite thrilled with the new features that the language is getting.
Ngscopeclient: Advanced T&M remote control and analysis suite (ngscopeclient.org)
Drag and drop to create complex, GPU-accelerated analysis pipelines in the filter graph editor
Slime mold simulation in Rust using WASM and WebGPU (github.com/plul)
Show HN: Datoviz – Vulkan-based GPU scientific visualization (C/C++/Python) (github.com/datoviz)
Pixi – rust-based package manager for reproducible scientific workflows (prefix.dev)
Neko: Portable framework for high-order spectral element flow simulations (github.com/ExtremeFLOW)
NumPy 2.0 Is Released (numpy.org)
Ndindex: A Python library for manipulating indices of ndarrays (quansight-labs.github.io)
ROOT: analyzing petabytes of data scientifically (root.cern)
LANL Achieves Yottabyte-Scale Data Compression in Neutron Transport Equations (hpcwire.com)
Fortran popularity rises with numerical and scientific computing (infoworld.com)
High performance array programming in Petalisp (zenodo.org)