|
Class Summary |
| DMROptimizable |
|
| DMRTopicModel |
|
| HierarchicalLDA |
|
| HierarchicalPAM |
Hierarchical PAM, where each node in the DAG has a distribution over all topics on the
next level and one additional "node-specific" topic. |
| LDA |
Deprecated. Use ParallelTopicModel instead. |
| LDAHyper |
Deprecated. Use ParallelTopicModel instead, which uses substantially faster data structures even for non-parallel operation. |
| LDAStream |
|
| MarginalProbEstimator |
An implementation of topic model marginal probability estimators
presented in Wallach et al., "Evaluation Methods for Topic Models", ICML (2009) |
| MultinomialHMM |
Latent Dirichlet Allocation. |
| NPTopicModel |
A non-parametric topic model that uses the "minimal path" assumption
to reduce bookkeeping. |
| PAM4L |
Four Level Pachinko Allocation with MLE learning,
based on Andrew's Latent Dirichlet Allocation. |
| ParallelTopicModel |
Simple parallel threaded implementation of LDA,
following the UCI NIPS paper, with SparseLDA
sampling scheme and data structure. |
| PolylingualTopicModel |
Latent Dirichlet Allocation for loosely parallel corpora in arbitrary languages |
| SimpleLDA |
A simple implementation of Latent Dirichlet Allocation using Gibbs sampling. |
| TopicalNGrams |
Like Latent Dirichlet Allocation, but with integrated phrase discovery. |
| TopicAssignment |
This class combines a sequence of observed features
with a sequence of hidden "labels". |
| TopicInferencer |
|
| WorkerRunnable |
A parallel topic model runnable task. |