search

lejon / PartiallyCollapsedLDA

Implementations of various fast parallelized samplers for LDA, including Partially Collapsed LDA, Light LDA, Partially Collapsed Light LDA and a very efficient Polya-Urn LDA
26 stars 20 forks source link

Issue in z files #25

Open MansMeg opened 1 year ago

MansMeg commented 1 year ago

Hi!

In the z_NNN.csv files there seem to be a bug with the pos variable. The file seem to be order by the vocabulary, but the pos is not corrected?

#doc source pos typeindex type topic
0 NA 0 0 access 21
0 NA 1 1 additional 1
0 NA 2 2 april 32
0 NA 3 3 authorize 27
0 NA 4 4 basis 32
0 NA 5 5 contract 32
0 NA 6 6 discharge 27
0 NA 7 7 doesnt 15
0 NA 8 8 expected 30
0 NA 9 9 fewer 27
0 NA 10 10 flow 32
0 NA 11 11 forecast 31
0 NA 12 12 half 32
0 NA 13 13 holtz 38
0 NA 14 14 imagination 38
0 NA 15 15 insight 49
0 NA 16 16 keith 9