DeepSpeed is a deep learning optimization library that makes distributed training and inference easy, efficient, and effective.

A library for accelerating Transformer models on NVIDIA GPUs, including using 8-bit and 4-bit floating point (FP8 and FP4) precision on Hopper, Ada and Blackwell GPUs, to provide better performance…

ncnn is a high-performance neural network inference framework optimized for the mobile platform

Extension module for Acoular for the realization of microphone array GUI applications with minimal programming effort.

Acoustic testing and source mapping software

TFLite Support is a toolkit that helps users to develop ML and deploy TFLite models onto mobile / ioT devices.

Learning in infinite dimension with neural operators.

Python based GBDT implementation on GPU. Efficient multioutput (multiclass/multilabel/multitask) training

A library for efficient similarity search and clustering of dense vectors.

Manipulation and analysis of geometric objects

Study resources for quantitative finance

Course on the ESP32 IDF

The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (V…

PyTorch image models, scripts, pretrained weights -- ResNet, ResNeXT, EfficientNet, EfficientNetV2, NFNet, Vision Transformer, MixNet, MobileNet-V3/V2, RegNet, DPN, CSPNet, and more

Fast and customizable framework for automatic ML model creation (AutoML)

Pen and paper exercises in machine learning

FFCV: Fast Forward Computer Vision (and other ML workloads!)

Fast and flexible image augmentation library. Paper about the library: https://www.mdpi.com/2078-2489/11/2/125

A library built upon PyTorch for building embeddings on discrete event sequences using self-supervision

Course about deep learning for computer vision and graphics co-developed by YSDA and Skoltech.

Differentiable ODE solvers with full GPU support and O(1)-memory backpropagation.

A short course in pseudospectral collocation methods for wave equations, with implementations in Python.

CuPy-compatible initial value problem library based on Scipy

Mamba SSM architecture

YSDA course in Natural Language Processing

GPU & Accelerator process monitoring for AMD, Apple, Huawei, Intel, NVIDIA and Qualcomm