designModelCorpora: designModelCorpora
In j2scode/predictifyR: Word Prediction Language Model Evaluation

Description Usage Arguments Details Value Author(s)

designModelCorpora Designs the model corpora

1 2	designModelCorpora(sampleSizeEstimate, registerSizeEstimate, samplingUnit, analysis)

`sampleSizeEstimate`	- the vocabulary-based estimate
`registerSizeEstimate`	- the lexical feature based estimate
`samplingUnit`	- the sampling unit
`analysis`	- the clean data analysis

This function takes as its parameters, the the vocabulary based sample size estimate from estimateSampleSize, the register size estimates from estimateRegisterSize, the sampling unit estimate from estimateSamplingUnit, and the clean data analysis from analyzeCorpus, and produces the model corpora design including sizes for four training sets, a validation set and a test set.

alpha - the corpus design for alpha training set.

John James

j2scode/predictifyR documentation built on May 14, 2019, 10:34 a.m.