lightonai Goto Github PK

.github

adversarial-robustness-by-design

A new defense mechanism against adversarial attacks through Optical Processing Units and synthetic gradients.

akronomicon

Public rankings of extreme-scale models

causal-conv1d-amd

Port of causal-conv1d to run and run efficiently on AMD.

chroma

the AI-native open-source embedding database

composer

Supercharge Your Model Training

contrastive-embeddings-for-neural-architectures

Architecture embeddings independent from the parametrization of the search space

datatrove

Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.

dfa-scales-to-modern-deep-learning

Study on the applicability of Direct Feedback Alignment to neural view synthesis, recommender systems, geometric learning, and natural language processing.

double-descent-curve

Double Descent Curve with Optical Random Features

double-trouble-in-double-descent

Double Trouble in the Double Descent Curve with Optical Processing Units.

js-lightonmuse

lairgpt

Inference code in Pytorch for GPT-like models, such as PAGnol, a family of models with up to 1.5B parameters, trained on datasets in French.

lightonml

Python library for running large-scale computations on LightOn's OPUs

lightonmuse

Python client for the LightOn Muse API

lightonmuse-sagemaker-sdk

mamba-amd

Port of Mamba to run and run efficiently on AMD.

muse-docs

Documentation for the Muse API.

muse-examples

Collection of examples for the Muse models.

muse-google-sheets

Source code for the integration of Muse in Google Sheets.

newma

Implementation of NEWMA: a new method for scalable model-free online change-point detection

newma-md

Conformational exploration SARS-CoV-2 (coronavirus responsible for COVID-19)

opu-benchmarks

ML benchmarks performance featuring LightOn's Optical Processing Unit (OPU) vs CPU and GPU.

outlines

Structured Text Generation

paradigm-client

Python client for LightOn Paradigm

phase-retrieval-opu

Minimal code to perform phase retrieval on LightOn OPUs

principled-dfa-training

Code for our paper on best practices to train neural networks with direct feedback alignment (DFA).

reinforcement-learning-opu

Code to perform Model-Free Episodic Control using Aurora OPUs

rita

RITA is a family of autoregressive protein models, developed by LightOn in collaboration with the OATML group at Oxford and the Debora Marks Lab at Harvard.

supervised-random-projections

Python implementation of supervised PCA, supervised random projections, and their kernel counterparts.