code/README.md

In nateaff/legolda: Color Topic Models of the Lego Dat Set

Code

This directory contains analysis code called by the .Rmd files in the analysis directory.

File contents, organized by the rmarkdown files which call chunks in the files and order of dependencies.

1. chunk-option: Global options from the rmarkdown files.

tf-idf.Rmd

1. tf-idf.R: TF-IDF analylsis and set plots

train-model.Rmd

1. perplexity-cv.R: 5-fold cross validation of perplexity scores
2. ldatuning-scores: Scores and plot based on method from the ldatuing package
3. coherence-scores: UMass internal coherence scores and plot
4. gamma-distributions: Distribution of sets over topic probabilitlies

topic-term-distributions.Rmd

1. compare-models.R: Plots of color distributions for select models and plot and tables of sets most associated with some topics.

final-model.Rmd

1. topic-names: Generate topic names for the final model
2. final-model-plots.R: Bar plot and time series of topic counts of the final topic model
3. final-model-grid.R: Waffle plots of the final topics

nateaff/legolda documentation built on May 18, 2019, 10:15 a.m.