Skip to contents

This function takes a trajectory model and reduces the number of motifs using 'prego'.

Usage

distill_traj_model(
  traj_model,
  max_motif_num,
  min_diff = 0.1,
  intra_cor_thresh = 0.6,
  use_non_linear = TRUE,
  parallel = TRUE
)

Arguments

traj_model

A trajectory model object

max_motif_num

The maximum number of motifs to select

min_diff

The minimum difference of scores to use for distillation

intra_cor_thresh

The threshold for the average intra-cluster correlation to split clusters. If NULL, no splitting is done.

use_non_linear

Whether to use non-linear features. If FALSE, a glm model without non-linear features is used.

parallel

Whether to use parallel processing

Value

A distilled trajectory model object with max_motif_num models + additional features.