Søren Kaae Sønderby

stronk. An easy to customize, strong type library with built in support for unit-like behavior

Kronos. Kronos: A Foundation Model for the Language of Financial Markets

lockfree_mpmc_queue. Lockfree, atomic, multi producer, multi consumer, C++, in process and inter-process queue

conditional-flow-matching. TorchCFM: a Conditional Flow Matching library

flow_matching. A PyTorch library for implementing flow matching algorithms, featuring continuous and discrete flow matching implementations. It includes practical examples for both text and image modalities.

modular. The Modular Platform (includes MAX & Mojo)

DataFrame. C++ DataFrame for statistical, financial, and ML analysis in modern C++

Cal-QL. official implementation for our paper Cal-QL: Calibrated Offline RL Pre-Training for Efficient Online Fine-Tuning (NeurIPS 2023)

CORL. High-quality single-file implementations of SOTA Offline and Offline-to-Online RL algorithms: AWAC, BC, CQL, DT, EDAC, IQL, SAC-N, TD3+BC, LB-SAC, SPOT, Cal-QL, ReBRAC

d3rlpy. An offline deep reinforcement learning library

seq. The seq library is a collection of original C++17 STL-like containers and related tools

aim. Aim 💫 — An easy-to-use & supercharged open-source experiment tracker.

static-frame. Immutable and statically-typeable DataFrames with runtime type and data validation

dataviz-in-python. A repository used to provide an introduction to dataviz in Python

zoo. Python

triton. Development repository for the Triton language and compiler

FLASH-pytorch. Implementation of the Transformer variant proposed in "Transformer Quality in Linear Time"

continual-transformers. Official Pytorch Implementation for "Continual Transformers: Redundancy-Free Attention for Online Inference" [ICLR 2023]

arrow2. Transmute-free Rust library to work with the Arrow format

alpaca. Serialization library written in C++17 - Pack C++ structs into a compact byte-array without any macros or boilerplate code

flash-attention. Fast and memory-efficient exact attention

mdspan. Reference implementation of mdspan targeting C++23

bde. Basic Development Environment - a set of foundational C++ libraries used at Bloomberg.

AnyAny. C++17 library for comfortable and efficient dynamic polymorphism

unordered_dense. A fast & densely stored hashmap and hashset based on robin-hood backward shift deletion

Pyraformer. Python

compile-time-regular-expressions. Compile Time Regular Expression in C++

rotograd. Official Pytorch's implementation of RotoGrad

pytorch-image-models. The largest collection of PyTorch image encoders / backbones. Including train, eval, inference, export scripts, and pretrained weights -- ResNet, ResNeXT, EfficientNet, NFNet, Vision Transformer (ViT), MobileNetV4, MobileNet-V3 & V2, RegNet, DPN, CSPNet, Swin Transformer, MaxViT, CoAtNet, ConvNeXt, and more

xformers. Hackable and optimized Transformers building blocks, supporting a composable construction.

power_perceiver. Machine learning experiments using the Perceiver IO model to forecast the electricity system (starting with solar)

top-github-users. Check your ranking in GitHub! Don't forget to star ⭐ this repository.

LibMTL. A PyTorch Library for Multi-Task Learning

arttime. arttime is a CLI application that blends beauty of ASCII / text art with functionality of clock / timer / pattern-based time manager in terminal ⏰

fiddle. Python

ml_collections. ML Collections is a library of Python Collections designed for ML use cases.

Transformers-Tutorials. This repository contains demos I made with the Transformers library by HuggingFace.

x-transformers. A concise but complete full-attention transformer with a set of promising experimental features from various papers

svector. Compact SVO optimized vector for C++17 or higher

fast-transformers. Pytorch library for fast transformer implementations

RWKV-LM. RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.

libassert. The most over-engineered C++ assertion library

quickcpplib. Eliminate all the tedious hassle when making state-of-the-art C++ 14 - 23 libraries!

nanobind. nanobind: tiny and efficient C++/Python bindings

templatedXML. Simple XML parser which uses compile-time customizable model-based approach to parse XML documents due to predefiend schema. Contains simple XDFX, FB2 parsers according to predefined models and provides usage examples

longformer. Longformer: The Long-Document Transformer

DeepSpeedExamples. Example models using DeepSpeed

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

min-decision-transformer. Minimal implementation of Decision Transformer: Reinforcement Learning via Sequence Modeling in PyTorch for mujoco control tasks in OpenAI gym

yggdrasil-decision-forests. A library to train, evaluate, interpret, and productionize decision forest models such as Random Forest and Gradient Boosted Decision Trees.

tink. Tink is a multi-language, cross-platform, open source library that provides cryptographic APIs that are secure, easy to use correctly, and hard(er) to misuse.

lwlog. Very fast synchronous and asynchronous C++17 logging library

mold. mold: A Modern Linker 🦠

nogil. Multithreaded Python without the GIL

memory. STL compatible C++ memory allocator library using a new RawAllocator concept that is similar to an Allocator but easier to use and write.

NanoLog. Nanolog is an extremely performant nanosecond scale logging system for C++ that exposes a simple printf-like API.

time_series_augmentation. An example of time series augmentation methods with Keras

awesome-cmake. A curated list of awesome CMake resources, scripts, modules and examples.

CudaRelativeAttention. custom cuda kernel for {2, 3}d relative attention with pytorch wrapper

pollnet. A collection of non-blocking(polling) network libs for Linux, also support solarflare APIs(Tcpdirect/Efvi)

str. A SIMD optimized fixed-length string class along with an adaptive hash table for fast searching

nanograd. A miniscule implementation of reverse mode auto-differentiation

A5. C++ Custom memory allocators

cute_headers. Collection of cross-platform one-file C/C++ libraries with no dependencies, primarily used for games

lib. single header libraries for C/C++

HandmadeMath. A simple math library for games and computer graphics. Compatible with both C and C++. Public domain and easy to modify.

poor-smime-sign. A poor mans tool to do S/MIME signatures in Python2 and 3.

libsodium. A modern, portable, easy to use crypto library.

cryptography. cryptography is a package designed to expose cryptographic primitives and recipes to Python developers.

pyopenssl. A Python wrapper around the OpenSSL library

Fast-Log. Toy fats log for Meeting C++ 2016

raster-vision. An open source library and framework for deep learning on satellite and aerial imagery.

cxx-std-draft. C++ standards drafts

OptTrees. Source code for: Nima Asadi, Jimmy Lin, and Arjen P. de Vries. Runtime Optimizations for Tree-Based Machine Learning Models. IEEE Transactions on Knowledge and Data Engineering, 26(9):2281-2292, 2014.

CCER. Objective-C

rapidstring. Copyright (c) 2018 John Boyer

AMQP-CPP. C++ library for asynchronous non-blocking communication with RabbitMQ

examples.amqp-cpp. AMQP-CPP examples based on libevent

SimpleAmqpClient. Simple C++ Interface to rabbitmq-c

forestpacking. C++

semimap. A semi compile-/run-time associative map container with compile-time lookup and run-time storage

ragel. Ragel State Machine Compiler -- MOVED TO colm-suite

linbin. Linear vs binary search benchmark

pitchfork. Pitchfork is a Set of C++ Project Conventions

xercesc. Parsing XML with Xerces-C++ example.

ClangBuildAnalyzer. Clang build analysis tool using -ftime-trace

array. contiguous container library - arrays with customizable allocation, small buffer optimization and more

type_safe. Zero overhead utilities for preventing bugs at compile time

SPSC_Queue. A highly optimized single producer single consumer message queue C++ template

SPSCQueue. A bounded single-producer single-consumer wait-free and lock-free queue written in C++11

vigra. a generic C++ library for image analysis

matplotlib-cpp. Extremely simple yet powerful header-only C++ plotting library built on the popular matplotlib

spdlog-cmake. sample cmake project with spdlog

OpenPCDet. OpenPCDet Toolbox for LiDAR-based 3D Object Detection.

PV-RCNN. PV-RCNN: Point-Voxel Feature Set Abstraction for 3D Object Detection, CVPR 2020.

carla. Open-source simulator for autonomous driving research.

PyMLViz. Interactive machine learning algorithm visualisation using Python Jupyter notebooks

TD-VAE. Temporal Difference Variational Auto-Encoder (TD-VAE) (Implementation in PyTorch)

namedtensor. Named Tensor implementation for Torch

ray. Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

dragan-pytorch. PyTorch implementation of DRAGAN (https://arxiv.org/abs/1705.07215)

Deep-learning-with-cats. Deep learning with cats (^._.^)

samplernn-pytorch. PyTorch implementation of SampleRNN: An Unconditional End-to-End Neural Audio Generation Model

pytorch-extension. an example of a CUDA extension for PyTorch using CuPy which computes the Hadamard product of two tensors

seq2seq.pytorch. Sequence-to-Sequence learning using PyTorch

tree_rnn. Theano implementation of Tree RNNs aka Recursive Neural Networks.

recurrent-batch-normalization. TeX

nninit. Weight initialisation schemes for Torch7 neural network modules

brainstorm. Fast, flexible and fun neural networks.

resume.github.com. Resumes generated using the GitHub informations

cgt. Computation Graph Toolkit

ladder. Ladder network is a deep learning algorithm that combines supervised and unsupervised learning

maxas. Assembler for NVIDIA Maxwell architecture

lasagne-draw. Implementation of the DRAW network in lasagne

mnist-cluttered. Cluttered MNIST Dataset

keras. Deep Learning for humans

research. Public research repository

rnn_ctc. Recurrent Neural Network and Long Short Term Memory (LSTM) with Connectionist Temporal Classification implemented in Theano. Includes a Toy training example.

nntools. Fork of https://github.com/Lasagne/Lasagne

Lasagne. Lightweight library to build and train neural networks in Theano

learning_to_execute. Learning to Execute

torch9. The ultimate Torch library.

DeepLearnToolbox. Matlab/Octave toolbox for deep learning. Includes Deep Belief Nets, Stacked Autoencoders, Convolutional Neural Nets, Convolutional Autoencoders and vanilla Neural Nets. Each method has examples to get you started.

DeepLearnToolbox. Matlab/Octave toolbox for deep learning. Includes Deep Belief Nets, Stacked Autoencoders, Convolutional Neural Nets, Convolutional Autoencoders and vanilla Neural Nets. Each method has examples to get you started.