Moses
statistical
machine translation
system
1. Moses
Overview
Manual
Online Demos
FAQ
Mailing Lists
Get Involved
Recent Changes
2. Getting Started
Source Installation
Baseline System
Packages
Releases
Sample Data
Links to Corpora
3. Tutorials
Phrase-Based Tutorial
Syntax Tutorial
Factored Tutorial
Optimizing Moses
Experiment.Perl
4. Training
Overview
Prepare training data
Factored Training
1
Prepare data
2
Run GIZA
3
Align words
4
Lexical translation
5
Extract phrases
6
Score phrases
7
Reordering model
8
Generation model
9
Configuration file
Language Models
Tuning
Training Reference
Decoder Reference
5.
User Documentation
Advanced Models
Efficient Phrase and Rule Storage
Search (Decoding)
Unknown Words
Hybrid Translation
Moses as a Service
Incremental Training
Domain Adaptation
Constrained Decoding
Cache-based Models
Sparse features
Support Tools
External Tools
Web Translation
Moses2
Pipeline Creation Language
Obsolete Features
6. Development
Video
Code Guide
Code Style
Factors
Feature Functions
Sparse Feature Functions
Code Documentation
Regression testing
7. Background
Statistical MT
SMT Glossary
Factored models
Confusion Networks
Word Lattices
Publications
Search
Components
»
PhraseTables
work in progress
Phrase table implementations for Moses
Moses includes
InMemory
only recommended for very small, filtered phrase tables (e.g., for tuning)
NOT recommended for any production system
OnDiskPT
the only implementation to support syntax-based MT
CompactPT
small and fast once built
filter phrase tables first for optimal performance
slow to build
Mmsapt
sampling phrase table based on suffix arrays
fast loads, fast performance due to caching
much faster built time than the other phrase tables; no phrase extraction or phrase scoring ahead of time required
the only implementation to support incremental updates
Edit
-
History
-
Print
Page last modified on October 26, 2015, at 10:24 PM