kentonmurray Goto Github PK

arabicdetokenizer

autograbscorers

Brittle and hacky way to grab some BLEU/Meteor/etc scorers

bevara

Android Phone Application designed for Linguistic Fieldwork to help preserve, maintain, and save endangered languages

buckwalter

A small python script that transliterates Arabic text using the Buckwalter Transliteration Scheme. It allows for multiple decisions to be made around whether or not to include all types of diacritics and characters or ignore them. Useful for NLP experiments where you may want to normalize text.

chameleon_translation

Machine Translation toolkits rely on numerous hyperparameters. Normally a user will not specify every possible value and toolkits will use default configurations. However, when attempting to replicate models and results, especially across different toolkits, this creates major issues. This project maps hyperparameters to a common yaml format and can read and generate scripts for numerous toolkits.

colbert

ColBERT: Contextualized Late Interaction over BERT (SIGIR'20)

computational_linguistic_analysis_2016

Code for a computational linguistic analysis of the 2016 presidential candidates. See www.kentonmurray.com/blogs/comp_linguistic_2016.html for an analysis.

craybook

Dat Shit Cray

curriculum_scorer

Scoring metrics for curriculum learning.

deepdomers

MTMA 2016 project using Tensor FLow

ducttape

A workflow management system for researchers who heart Unix.

fairseq-for-en-ar

farasapostprocessor

This takes the FARASA outputs and using heuristics combines certain split words back together to simplify tasks for human annotators

git-ftp

Uses Git to upload only changed files to FTP servers.

hw0-kwmurray

Stupid Assignment for a Software Engineering Class

hw1-kwmurray

Midterm for Software Engineering Class

hw2-team06

Final project (HW 2) for a software engineering class

hyde

introhlt

These are the teaching materials that I created for Introduction to Human Language Technology (EN. 601.647/667) at JHU. This is taught by multiple people so this only some of the lectures.

java

Java related code

just

A tool for constructing and executing experiment pipelines on a cluster

kentonsbibtex

This is just a personal BibTex file of papers that I may occasionally use in papers. It is by no means inclusive, nor meant for others, but still feel free to pull, though I likely won't accept push requests

keras-rl

Deep Reinforcement Learning for Keras.

lamtram

lamtram: A toolkit for neural language and translation modeling

lightning-transformers

Flexible interface for high-performance research using SOTA Transformers leveraging Pytorch Lightning, Transformers, and Hydra.

lstm

mascsll.github.io

2024 MASC-SLL website

mastodon_mt_data

Code to get posts from the mastodon feed for MT purposes

mosesdecoder

Moses, the machine translation system

mtdata

A tool that locates, downloads, and extracts machine translation corpora