Sebastian Vincent's repositories
fishWaffle
A text generation system based on a LSTM-powered language model trained on k-punk's blog posts
os_cxt_extractor
A Python module used to extract pairs of consecutive sentences parallel in two languages from the OpenSubtitles18 corpus.
8-puzzle-search
8-puzzle search implementation using BFS and A*
abduction_engine
An implementation for a probabilistic abduction engine, with examples embedded into the GUI.
algorithms
My universal implementations of algorithms in C++/python, all in one place
aoc2021
My solutions to Advent of Code 2021 puzzles in Python 3.
cxt2vec
Software for vectorising contexts (such as metadata) to use in models like MTCue.
eamt22_evaluation
Repository containing code and data necessary to evaluate approaches to the EAMT22 English-to-Polish translation task.
fairseq
Facebook AI Research Sequence-to-Sequence Toolkit written in Python.
cornell_rich
Cornell-rich: a dataset of rich speaker annotations + film metadata for the popular Cornell Movie Dialogs Corpus.