John Thickstun

A beautiful, simple, clean, and responsive Jekyll theme for academics

Anticipatory Autoregressive Models

Implementation of the BASIS algorithm for source separation with deep generative priors

PyTorch DataSet and Jupyter demos for MusicNet

Arithmetic coding for long-context modeling

Code for watermarking language models

371 Four-part Chorales by J.S. Bach in the Humdrum file format.

Legibile, Scalable, Reproducible Foundation Models with Named Tensors and Jax

Homework 2 for Generative Models

Homework 3 for Generative Models

Homework 1 for Generative Models

The mini-MusicNet dataset

Experiments for Invariances and Data Augmentation for Supervised Music Transcription

Some experiments with the Lean proof assistant

Dataset and evaluation pipeline for music-to-score alignment

Experiments for Learning Features of Music From Scratch, ICLR 2017

Experiments for Coupled Recurrent Networks for Polyphonic Music Composition

Longformer: The Long-Document Transformer

Mistral: A strong, northwesterly wind: Framework for transparent and accessible large-scale language model training, built with Hugging Face 🤗 Transformers.

Music of Gaspar van Weerbeke as Humdrum digital scores.

🤗 The largest hub of ready-to-use datasets for ML models with fast, easy-to-use and efficient data manipulation tools

🤗Transformers: State-of-the-art Natural Language Processing for Pytorch and TensorFlow 2.0.

Code for reproducing results in "Glow: Generative Flow with Invertible 1x1 Convolutions"

Digital edition of Joseph Haydn's string quartets in the Humdrum file format.

Digital edition of Frédéric Chopin's mazurkas in the Humdrum file format.

Music of Antoine Busnoys as Humdrum digital scores.

Music of Scott Joplin in the Humdrum file format.

Digital edition of W.A. Mozart's string quartets in the Humdrum file format.

Digital edition of Domenico Scarlatti's keyboard sonatas in the Humdrum file format

Digital edition of Johann Nepomuk Hummel's preludes in the Humdrum file format.