wlearn is an open-source classical ML toolkit for JavaScript: boosted trees, random forests, GAMs, SVMs, clustering, nearest neighbors and more, powered by WebAssembly and native C11 implementations. wlearn.org/
Just released polygrad 0.4.2: a portable tensor engine for Python, Node.js and browsers. It's a C11 port of tinygrad's core with Python/JS frontends, WASM/WebGPU/CUDA/CPU/X86 backends, plus exportable artifacts. Available on npm and PyPi as polygrad.
github.com/polygrad/polygrad
StatSim Profile update: new data profiling core for faster streaming stats, browser profiling now runs in a Web Worker to keep the UI responsive on large files, plus URL-based data loading and CLI tooling. Try: github.com/statsim/profile
🎉 We're back! After a few years of inactivity, we're excited to announce the release of StatSim 0.13.2. Featuring a flow-based model editor and a revamped user interface. As before, it's fully open-source and free. Check the latest version on statsim.com/app
StatSim 0.9.0 is out featuring units, parallel simulations and numerous bug fixes. You can now add units to your data, "guess" units automatically in expressions, run the same model using multiple chains simultaneously.
Web: statsim.com/app/
Download: github.com/statsim/app/relea… (Windows, Linux)
StatSim 0.8.0 is now available, featuring a reactive table editor for model data, probability density charts (KDE) and performance improvements.
StatSim 0.7.0 is now available, bringing neural nets to your models! Play with multilayer neural networks in your browser. Compile StatSim projects to html files or system binaries using server-side processing. Couple of new UI and architectural changes are now available. #ML
StatSim 0.6.1 is out, featuring numerous bug fixes, MAP estimates, descriptive statistics, autocovariograms. You can also download binaries for Windows and Linux to use StatSim offline!
github.com/statsim/app/relea…













