Sequence modeling¶
Language models¶
Stores language model vocabulary. |
|
Class for counting ngrams. |
|
Class for providing MLE ngram model scores. |
|
Provides Lidstone-smoothed scores. |
|
Implements Laplace (add one) smoothing. |
|
Interpolated version of Witten-Bell smoothing. |
|
Interpolated version of Kneser-Ney smoothing. |
|
Interpolated version of smoothing with absolute discount. |
|
Provides StupidBackoff scores. |
Translation¶
Return an aligned sentence object, which encapsulates two sentences along with an |
|
A storage class for representing alignment between two sequences, s1, s2. |
|
In-memory store of translations for a given phrase, and the log probability of the those translations |
|
Abstract base class for all IBM models |
|
Lexical translation model that ignores word order |
|
Lexical translation model that considers word order |
|
Translation model that considers how a word can be aligned to multiple words in another language |
|
Translation model that reorders output words based on their type and their distance from other related words in the output sentence |
|
Translation model that keeps track of vacant positions in the target sentence to decide where to place translated words |
|
Phrase-based stack decoder for machine translation |
|
|
Traverse the alignment cost from the tracebacks and retrieves appropriate sentence pairs. |
|
This module symmetrisatizes the source-to-target and target-to-source word alignment output and produces, aka. |
|
This function checks for alignment point consistency and extracts phrases using the chunk of consistent phrases. |