| Package | Description |
|---|---|
| com.aliasi.hmm |
Classes for estimating and decoding hidden Markov models.
|
| Modifier and Type | Class and Description |
|---|---|
class |
AbstractHmm
An
AbstractHmm is an abstract implementation of a
hidden Markov model which manages a symbol table and defines the
basic methods in terms of the symbolic linear probability methods. |
class |
AbstractHmmEstimator
An
HmmEstimator may be used to train a hidden Markov
model (HMM). |
class |
HmmCharLmEstimator
An
HmmCharLmEstimator employs a maximum a posteriori
transition estimator and a bounded character language model
emission estimator. |
| Modifier and Type | Method and Description |
|---|---|
HiddenMarkovModel |
HmmDecoder.getHmm()
Returns the hidden Markov model underlying this decoder.
|
| Constructor and Description |
|---|
HmmDecoder(HiddenMarkovModel hmm)
Construct an HMM decoder from the specified HMM.
|
HmmDecoder(HiddenMarkovModel hmm,
Map<String,double[]> emissionCache,
Map<String,double[]> emissionLog2Cache)
Construct an HMM decoder from the specified HMM using the
specified caches for linear and log probabilities.
|
HmmDecoder(HiddenMarkovModel hmm,
Map<String,double[]> emissionCache,
Map<String,double[]> emissionLog2Cache,
double log2Beam,
double log2EmissionBeam)
Construct an HMM decoder from the specified HMM using the
specified caches for linear and log probabilities, with the
specified beam width for emission estimates.
|
Copyright © 2019 Alias-i, Inc.. All rights reserved.