Official inference framework for 1-bit LLMs

A functional standard library for Python.

Model2Vec: Distill a Small Fast Model from any Sentence Transformer

Open source libraries and APIs to build custom preprocessing pipelines for labeling, training, or production machine learning pipelines.

Code for explaining and evaluating late chunking (chunked pooling)

A compute framework for building Search, RAG, Recommendations and Analytics over complex (structured+unstructured) data, with ultra-modal vector embeddings.

Machine Learning Engineering Open Book

Summaries and resources for Designing Machine Learning Systems book (Chip Huyen, O'Reilly 2022)

YSDA course in Natural Language Processing

Kedro is a toolbox for production-ready data science. It uses software engineering best practices to help you create data engineering and data science pipelines that are reproducible, maintainable,…

The lazier way to manage everything docker

simple terminal UI for git commands

Publication-ready NN-architecture schematics.

On the Validity of Using the Therapist's prompts in Automatic Depression Detection from Clinical Interview

Official source code for the paper: "Reading Between the Frames Multi-Modal Non-Verbal Depression Detection in Videos"

Track emissions from Compute and recommend ways to reduce their impact on the environment.

ML has an impact on the climate. But not all models are born equal. Compute your model's emissions with our calculator and add the results to your paper with our generated latex template

A Python library for performing NLP on clinical text written in Dutch

Repository for the textbook 'Improving Your Statistical Inferences' by Daniel Lakens

Representation learning on dynamic graphs using self-attention networks

mixup: Beyond Empirical Risk Minimization

Video+code lecture on building nanoGPT from scratch

Fast and memory-efficient exact attention

PixMix: Dreamlike Pictures Comprehensively Improve Safety Measures (CVPR 2022)

Weaviate is an open-source vector database that stores both objects and vectors, allowing for the combination of vector search with structured filtering with the fault tolerance and scalability of …

Code for Deep-learning Architecture for Short-term Passenger Flow Forecasting in Urban Rail Transit

Code for the paper: "No Zero-Shot Without Exponential Data: Pretraining Concept Frequency Determines Multimodal Model Performance" [NeurIPS'24]

Kolmogorov Arnold Networks

Implementation of "NITI: Training Integer Neural Networks Using Integer-only Arithmetic" on arxiv

A minimal GPU design in Verilog to learn how GPUs work from the ground up