moskomule

dda

Differentiable Data Augmentation Library

homura

homura is a library for fast prototyping DL research

anatome

Ἀνατομή is a PyTorch library to analyze representation of neural networks

chika

chika is a simple and easy config tool for hierarchical configurations.

hypergrad

Simple and extensible hypergradient for PyTorch

memory_efficient_attention.pytorch

A human-readable PyTorch implementation of "Self-attention Does Not Need O(n^2) Memory" (Rabe&Staats'21).

dataset-contamination

Datasets of "Will Large-scale Generative Models Corrupt Future Datasets?"

matplotlib-better-practice

moskomule.github.io

pyproject_template

fast_dataloader

Fast data loader using tar and tensordict

iclr

In Context Learning Research

LaTeX-better-practice

To avoid re-searching and re-inventing wheels

moskomule

chatgpt

dotfiles

edm

Elucidating the Design Space of Diffusion-Based Generative Models (EDM)

equinox_vision

Vision models with equinox

evojax

materials

Materials for slides

mkposters

Make posters from Markdown files.

mlx

MLX: An array framework for Apple silicon

mlx-examples

Examples in the MLX framework

mlx_examples

MNISQ-quantum-circuit-dataset

MNISQ circuit dataset for machine learning and quantum machine learning

project-template

This is a template repository for my research projects.

redirect

stable-diffusion

A latent text-to-image diffusion model

sweeper

hyperparameter sweeper

symplectic-adjoint-method

Code for "Symplectic Adjoint Method for Exact Gradient of Neural ODE with Minimal Memory," NeurIPS, 2021.