SALM: Suffix Array and its Applications in Empirical Language Processing by Joy
Public repository record indexed from GitHub. Explore verified star velocity metrics, source code repositories, and curated developer tool directories across the GitHubRepo ecosystem.
GIZA++ is a statistical machine translation toolkit that is used to train IBM Models 1-5 and an HMM word alignment model. This package also contains the source for the mkcls tool which generates the word classes necessary for training some of the alignment models.
Frozen version of Vowpal Wabbit for use with Moses.
A word alignment tool based on famous GIZA++, extended to support multi-threading, resume training and incremental training.
Regression test data and test cases