larspars

RWKV-LM

RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.

crux-top-lists

Downloadable snapshots of the Chrome Top Million Websites pulled from public CrUX data in Google BigQuery.

deep-thinking

A centralized place for deep thinking code and experiments

bolt

10x faster matrix and vector operations

CogView2

official code repo for paper "CogView2: Faster and Better Text-to-Image Generation via Hierarchical Transformers"

PhotonLibOS

Probably the fastest coroutine lib in the world!

financial-charts

Stock Asset Allocation vs Annualized 10-year SPX Return. (Website)

ClickHouse.Client

.NET client for ClickHouse

html2canvas

Screenshots with JavaScript

einops

Flexible and powerful tensor operations for readable and reliable code (for pytorch, jax, TF and others)

scikit-shape

Python package for shape optimization and shape analysis in image and data processing

2020-instanceSeg

Deep Variational Instance Segmentation

programmingishard

Long-term book project

Burstprop

Code to train on MNIST, CIFAR-10 and ImageNet using burstprop.

AttackVectors

A repository to monitor attack vectors from state-backed information operations

colmap

COLMAP - Structure-from-Motion and Multi-View Stereo

geomloss

Geometric loss functions between point clouds, images and volumes

nvdbapi-V2

Python-rutine for å vegnett og fagdata fra Nasjonal Vegdatabank API

carla

Open-source simulator for autonomous driving research.

pytorch-kaldi

pytorch-kaldi is a project for developing state-of-the-art DNN/RNN hybrid speech recognition systems. The DNN part is managed by pytorch, while feature extraction, label computation, and decoding are performed with the kaldi toolkit.

local-loss

PyTorch code for training neural networks without global back-propagation

TensorComprehensions

A domain specific language to express machine learning workloads.

transparent_latent_gan

Use supervised learning to illuminate the latent space of GAN for controlled generation and edit

fairseq

Facebook AI Research Sequence-to-Sequence Toolkit written in Python.

ToolsOfTheTrade

Tools of The Trade, from Hacker News.

vid2vid

Pytorch implementation of our method for high-resolution (e.g. 2048x1024) photorealistic video-to-video translation.

russian-troll-tweets

delta_orthogonal_init

Delta orthogonal initialization for Lua Torch

javascript-error-logging

A collection of JavaScript error logging services, resources and shiny things.

resnet-brelu

ResNet and DenseNet with Bipolar Rectified Linear Units (BReLU)