LDABatch: LDA Replications on a Batch System
In JonasRieger/ldaPrototype: Prototype of Multiple Latent Dirichlet Allocation Runs

LDABatch

R Documentation

LDA Replications on a Batch System

Description

Performs multiple runs of Latent Dirichlet Allocation on a batch system using the batchtools-package.

Usage

LDABatch(
  docs,
  vocab,
  n = 100,
  seeds,
  id = "LDABatch",
  load = FALSE,
  chunk.size = 1,
  resources,
  ...
)

Arguments

`docs`	[`list`] Documents as received from `LDAprep`.
`vocab`	[`character`] Vocabularies passed to `lda.collapsed.gibbs.sampler`. For additional (and necessary) arguments passed, see ellipsis (three-dot argument).
`n`	[`integer(1)`] Number of Replications.
`seeds`	[`integer(n)`] Random Seeds for each Replication.
`id`	[`character(1)`] Name for the registry's folder.
`load`	[`logical(1)`] If a folder with name `id` exists: should the existing registry be loaded?
`chunk.size`	[`integer(1)`] Requested chunk size for each single chunk. See `chunk`.
`resources`	[`named list`] Computational resources for the jobs to submit. See `submitJobs`.
`...`	additional arguments passed to `lda.collapsed.gibbs.sampler`. Arguments will be coerced to a vector of length `n`. Default parameters are `alpha = eta = 1/K` and `num.iterations = 200`. There is no default for `K`.

Details

The function generates multiple LDA runs with the possibility of using a batch system. The integration is done by the batchtools-package. After all jobs of the corresponding registry are terminated, the whole registry can be ported to your local computer for further analysis.

The function returns a LDABatch object. You can receive results and all other elements of this object with getter functions (see getJob).

Value

[named list] with entries id for the registry's folder name, jobs for the submitted jobs' ids and its parameter settings and reg for the registry itself.

Examples

## Not run: 
batch = LDABatch(docs = reuters_docs, vocab = reuters_vocab, n = 4, K = 15)
batch
getRegistry(batch)
getJob(batch)
getLDA(batch, 2)

batch2 = LDABatch(docs = reuters_docs, vocab = reuters_vocab, K = 15, chunk.size = 20)
batch2
head(getJob(batch2))

## End(Not run)

JonasRieger/ldaPrototype documentation built on Feb. 5, 2023, 6:45 p.m.