search

juliasilge / tidytext

Text mining using tidy tools :sparkles::page_facing_up::sparkles:
https://juliasilge.github.io/tidytext/
Other
1.18k stars 184 forks source link

Implement tidiers for topic models from stm package #51

Closed juliasilge closed 7 years ago

juliasilge commented 7 years ago

Let's write tidiers for the topic models from stm. This package has no rJava dependency and the same input as the topicmodels LDA modeling.

cschwem2er commented 7 years ago

Hi, is this currently in development? I'd love to combine tidytext and stm for teaching in mid October and a tidier function would be most useful for this.

juliasilge commented 7 years ago

I worked on this some today and it is almost good to go! Should be ready very shortly, in time for your course.

cschwem2er commented 7 years ago

that's awesome, thank you so much for this :)

juliasilge commented 7 years ago

@methodds Got this finished up this evening. Give it a go and please let us know if you run into any problems!

cschwem2er commented 7 years ago

Hi, I just tested the latest version and it works without any issues, again thank you so much for implementing this! :)

There's only one suggestion that I have for your example in stm_tidiers: STM offers the spectral method for model initialization which, unlike the usual initialization for LDA models, is deterministic (more details on that in the STM vignette). The spectral method is highly recommended by the STM authors. For this reason I think you should remove the seed, set.seed(2016) and just slightly adjust the stm call to topic_model <- stm(inaug, K = 3, verbose = FALSE, init.type = 'Spectral').

cschwem2er commented 7 years ago

Sorry just another quick question related to this: Do you have a time schedule for the next CRAN release? I dunno about other countries, but university courses in Germany start mid/end October and getting devtools to run in our labs is a real pain.

juliasilge commented 7 years ago

@methodds I will see what we can do about a quick-ish CRAN release! I'd like to get the tidyeval stuff all worked out for all the functions.

juliasilge commented 7 years ago

@methodds Just a heads up that the version of tidytext with stm tidiers is on CRAN now.

github-actions[bot] commented 2 years ago

This issue has been automatically locked. If you believe you have found a related problem, please file a new issue (with a reprex: https://reprex.tidyverse.org) and link to this issue.