Skip to content

Instantly share code, notes, and snippets.

@wleoncio

wleoncio/analysis.R

Last active November 19, 2025 13:45
Show Gist options

  • Select an option

    Learn more about clone URLs
  • Save wleoncio/8e073aac063ed72fb1025e6f6c71d79f to your computer and use it in GitHub Desktop.

Select an option

Learn more about clone URLs
Save wleoncio/8e073aac063ed72fb1025e6f6c71d79f to your computer and use it in GitHub Desktop.
Download ZIP
Error from BayesMallows::assign_cluster()
Raw
analysis.R
library(BayesMallows)
library(tidyverse)
library(parallel)
#load our data
data <- read.csv("data.csv")
#we reshape it to fit BayesMallows package
flavours <- c(
"Banan", "Caramel", "Chocolate Toffee", "Cocos", "Daim",
"Eclairs", "Fransk nougat", "Golden Toffee", "Japp", "Lakris",
"Marsipan", "Nougatcrisp", "NC8tti"
)
data_num <- apply(data[5:17], 2, function(x) match(x, flavours)) #replace characters with numbers
data_num_ord <- create_ordering(data_num) #replace ranking with ordering
colnames(data_num_ord) <- flavours #colnames and rownames just to remember
rownames(data_num_ord) <- paste0("Assessor", 1:dim(data_num)[1]) #colnames and rownames just to remember
data_BM <- setup_rank_data(rankings = data_num_ord)
#step 1 - observe and select burning
cl <- makeCluster(detectCores())
bmm <- compute_mallows_mixtures(
n_clusters = c(1, 2, 3, 4),
data = data_BM,
compute_options = set_compute_options(nmc = 10000, include_wcd = FALSE),
cl = cl)
stopCluster(cl)
assess_convergence(bmm)
selected_burnin <- 5000L
thinning <- 10L
#step 2 - observe and select the number of clusters
cl <- makeCluster(detectCores())
bmm <- compute_mallows_mixtures(
n_clusters = 1:4,
data = data_BM,
compute_options =
set_compute_options(nmc = 11000, burnin = selected_burnin,
rho_thinning = thinning, include_wcd = TRUE),
cl = cl)
stopCluster(cl)
plot_elbow(bmm)
selected_n_clusters <- 3
#step 3 - compute clustering and see the results
bmm <- compute_mallows(
data = data_BM,
model_options = set_model_options(n_cluster = selected_n_clusters),
compute_options = set_compute_options(
nmc = 11000, burnin = selected_burnin,
clus_thinning = thinning, rho_thinning = thinning, include_wcd = TRUE) #save_ind_clus = TRUE
)
assign_cluster(bmm)
Raw
data.csv
Time Gender Generation Time in Norway Rank1 Rank2 Rank3 Rank4 Rank5 Rank6 Rank7 Rank8 Rank9 Rank10 Rank11 Rank12 Rank13
2025-11-17 19:18:41 Female Gen Z >15 years Lakris Eclairs Marsipan Caramel Banan Japp Daim Notti
2025-11-17 19:25:22 Male Boomer >15 years Lakris Cocos Japp Marsipan Daim Caramel Notti
2025-11-17 19:28:10 Female Generation X >15 years Eclairs Caramel Lakris Marsipan Notti Japp
2025-11-17 19:28:38 Male Gen Z >15 years Caramel Lakris Cocos Marsipan Notti Daim Japp
2025-11-17 19:55:12 Female Generation X >15 years Banan Lakris Eclairs Japp Marsipan Daim Notti Caramel
Male Millenial >15 years Cocos Japp Marsipan Lakris Notti Caramel
Female Gen Z >15 years Lakris Marsipan Eclairs Japp Notti
Male Boomer >15 years Marsipan Cocos Lakris Daim Japp Notti
Female Generation X >15 years Eclairs Lakris Marsipan Caramel Japp Notti
Male Gen Z >15 years Lakris Japp Marsipan Cocos Caramel Daim Notti
Female Generation X <5 years Caramel Eclairs Japp Notti Marsipan Lakris
Male Millenial <5 years Daim Japp Cocos Caramel Banan Notti Marsipan Lakris
Female Gen Z <5 years Daim Eclairs Caramel Notti Japp Marsipan Lakris
Male Boomer <5 years Cocos Caramel Japp Notti Lakris Marsipan
Female Generation X <5 years Eclairs Daim Japp Notti Lakris Caramel
Male Gen Z <5 years Cocos Japp Banan Daim Notti Marsipan Lakris Caramel
Female Generation X <5 years Eclairs Notti Japp Caramel Banan Lakris Marsipan
Male Millenial <5 years Japp Cocos Daim Notti Caramel Lakris Marsipan
Female Generation X <5 years Daim Notti Eclairs Japp Caramel Marsipan Lakris
Male Millenial <5 years Cocos Banan Notti Daim Caramel Lakris Marsipan Japp
Female Gen Z >15 years Lakris Marsipan Eclairs Japp Caramel Daim
Male Boomer >15 years Marsipan Lakris Cocos Caramel Banan Daim
Female Generation X >15 years Eclairs Lakris Marsipan Caramel Daim
Male Gen Z >15 years Cocos Lakris Marsipan Banan Caramel Daim
Female Generation X >15 years Lakris Marsipan Eclairs Banan Caramel Daim
Male Millenial >15 years Lakris Marsipan Banan Cocos Caramel
Female Gen Z >15 years Eclairs Lakris Marsipan Banan Daim Caramel
Male Boomer >15 years Marsipan Cocos Lakris Banan Daim Caramel
Female Generation X >15 years Marsipan Lakris Banan Eclairs Daim Caramel
Male Gen Z >15 years Lakris Marsipan Cocos Banan Daim Caramel
Female Generation X <5 years Eclairs Banan Daim Caramel Lakris Marsipan
Male Millenial <5 years Japp Cocos Banan Caramel Daim Marsipan Lakris
Female Gen Z <5 years Eclairs Japp Banan Caramel Marsipan Lakris
Male Boomer <5 years Cocos Banan Caramel Daim Lakris
Female Generation X <5 years Eclairs Banan Caramel Daim Lakris Marsipan
Male Gen Z <5 years Cocos Caramel Japp Daim Lakris Marsipan
Female Generation X <5 years Eclairs Caramel Banan Daim Lakris Marsipan
Male Millenial <5 years Cocos Banan Caramel Japp Daim Marsipan Lakris
Female Generation X <5 years Eclairs Caramel Daim Lakris Marsipan
Male Millenial <5 years Cocos Caramel Japp Lakris Marsipan
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment