TGL kmeans with 'tidy' output
TGL_kmeans_tidy(
df,
k,
metric = "euclid",
max_iter = 40,
min_delta = 0.0001,
verbose = FALSE,
keep_log = FALSE,
id_column = FALSE,
reorder_func = "hclust",
add_to_data = FALSE,
hclust_intra_clusters = FALSE,
seed = NULL,
use_cpp_random = FALSE
)a data frame or a matrix. Each row is a single observation and each column is a dimension. the first column can contain id for each observation (if id_column is TRUE), otherwise the rownames are used.
number of clusters. Note that in some cases the algorithm might return fewer clusters than k.
distance metric for kmeans++ seeding. can be 'euclid', 'pearson' or 'spearman'
maximal number of iterations
minimal change in assignments (fraction out of all observations) to continue iterating
display algorithm messages
keep algorithm messages in 'log' field
df's first column contains the observation id. If not set and the first
column is character or factor, it will be automatically used as the ID column (with a warning).
function to reorder the clusters. operates on each center and orders by the result. e.g. reorder_func = mean would calculate the mean of each center and then would reorder the clusters accordingly. If reorder_func = hclust the centers would be ordered by hclust of the euclidean distance of the correlation matrix, i.e. hclust(dist(cor(t(centers))))
if NULL, no reordering would be done.
return also the original data frame with an extra 'clust' column with the cluster ids ('id' is the first column)
run hierarchical clustering within each cluster and return an ordering of the observations.
seed for the c++ random number generator
use c++ random number generator instead of R's. This should be used for only for backwards compatibility, as from version 0.4.0 onwards the default random number generator was changed to R.
list with the following components:
tibble with `id` column with the observation id (`1:n` if no id column was supplied), and `clust` column with the observation assigned cluster.
tibble with `clust` column and the cluster centers.
tibble with `clust` column and `n` column with the number of points in each cluster.
tibble with `clust` column the original data frame.
messages from the algorithm run (only if keep_log = TRUE).
tibble with 'id' column, 'clust' column, 'order' column with a new ordering if the observations and 'intra_clust_order' column with the order within each cluster. (only if hclust_intra_clusters = TRUE)
# create 5 clusters normally distributed around 1:5
d <- simulate_data(
n = 100,
sd = 0.3,
nclust = 5,
dims = 2,
add_true_clust = FALSE,
id_column = FALSE
)
head(d)
#> V1 V2
#> 1 0.4162643 1.0408562
#> 2 0.8653655 1.3884966
#> 3 1.0570451 0.6128050
#> 4 0.5543186 0.7068720
#> 5 0.5403417 1.4741502
#> 6 0.8346604 0.9403844
# cluster
km <- TGL_kmeans_tidy(d, k = 5, "euclid", verbose = TRUE)
#> will generate seeds
#> generating seeds
#> at seed 0
#> add new core from 97 to 0
#> at seed 1
#> done update min distance
#> seed range 350 450
#> picked up 410
#> add new core from 410 to 1
#> at seed 2
#> done update min distance
#> seed range 300 400
#> picked up 221
#> add new core from 221 to 2
#> at seed 3
#> done update min distance
#> seed range 250 350
#> picked up 175
#> add new core from 175 to 3
#> at seed 4
#> done update min distance
#> seed range 200 300
#> picked up 339
#> add new core from 339 to 4
#> reassign after init
#> iter 0
#> iter 1 changed 14
#> iter 1
#> iter 2 changed 2
#> iter 2
#> iter 3 changed 0
km
#> $centers
#> # A tibble: 5 × 3
#> clust V1 V2
#> <int> <dbl> <dbl>
#> 1 1 1.98 2.00
#> 2 2 3.98 3.97
#> 3 3 3.07 3.01
#> 4 4 0.983 0.980
#> 5 5 5.03 4.98
#>
#> $cluster
#> # A tibble: 500 × 2
#> id clust
#> <chr> <int>
#> 1 1 4
#> 2 2 4
#> 3 3 4
#> 4 4 4
#> 5 5 4
#> 6 6 4
#> 7 7 4
#> 8 8 4
#> 9 9 4
#> 10 10 4
#> # ℹ 490 more rows
#>
#> $size
#> # A tibble: 5 × 2
#> clust n
#> <int> <int>
#> 1 1 104
#> 2 2 99
#> 3 3 98
#> 4 4 99
#> 5 5 100
#>
#> $metric
#> [1] "euclid"
#>
#> attr(,"class")
#> [1] "tgl_kmeans" "list"