Software & Datasets

Here is a list of open-source code (& datasets) related to my projects in statistics, machine learning, and natural language processing over the years.

All public repositories can also be found on my GitHub page.

Python Packages on PyPI

• comparecast: Package for comparing sequential forecasters using confidence sequences & e-processes. Code accompanying our OR'23 paper.
• word2word: Easy-to-use word-to-word translations for 3,564 language pairs. Code accompanying our LREC'20 paper.

Selected Repositories on GitHub

• CombiningEvidenceAcrossFiltrations: Python code accompanying our preprint on combining evidence across filtrations.
• ComparingAbstainingClassifiers: Python code from our NeurIPS'23 paper.
• irm-empirical-study: Python code and data from our ICML'20 workshop paper.
• KorNLUDatasets: Korean NLI and STS datasets from our EMNLP-F'20 paper.
• helo_word: Python code for grammatical error correction (GEC), accompanying our BEA'19 workshop paper.
• deep_learning: An (old) Python implementation of standard deep neural network architectures and learning algorithms from scratch, as part of a deep learning course at CMU.