Efficiently computes derivatives of NumPy code.

Curated list of datasets and tools for post-training.

Simple RL training for reasoning

GitHub has sold us out. Time to get out.

Fast, flexible and easy to use probabilistic modelling in Python.

PyTorch implementation of TabNet paper : https://arxiv.org/pdf/1908.07442.pdf

NeMo Retriever Library is a scalable, performance-oriented document content and metadata extraction microservice. NeMo Retriever extraction uses specialized NVIDIA NIM microservices to find, contex…

A comprehensive set of fairness metrics for datasets and machine learning models, explanations for these metrics, and algorithms to mitigate bias in datasets and models.

Closed-form Continuous-time Neural Networks

LLM KV cache compression made easy

Leave One Feature Out Importance

Adan: Adaptive Nesterov Momentum Algorithm for Faster Optimizing Deep Models

Implementation of Rotary Embeddings, from the Roformer paper, in Pytorch

Fast Differentiable Sorting and Ranking

For optimization algorithm research and development.

NVIDIA Data Science stack tools

A collection of LogitsProcessors to customize and enhance LLM behavior for specific tasks.

Testing baseline LLMs performance across various models

XGBoost for label-imbalanced data: XGBoost with weighted and focal loss functions

RAdam implemented in Keras & TensorFlow

Quick implementation of nGPT, learning entirely on the hypersphere, from NvidiaAI

A tool to configure, launch and manage your machine learning experiments.

Our solution for the arc challenge 2024

The AdEMAMix Optimizer: Better, Faster, Older.

Unofficial implementation of Tiny Recursive Model (TRM), improvement to HRM from Sapient AI, by Alexia Jolicoeur-Martineau

NVARC solution to ARC-AGI-2

Create an AI capable of solving reasoning tasks it has never seen before