Plots a single sequence of observation.

plot_hsmm_seq(
  X,
  model,
  title = NULL,
  show_state_diff = TRUE,
  compact_view = FALSE,
  add_color_legend_in_compact_view = TRUE,
  add_state_color_legend = FALSE,
  selection = data.frame(),
  verbose = FALSE
)

Arguments

X	a `data.frame` specifying the sequence of observation. Should contains data for only one sequence. Each column is a variable. Any column starting with the characters `state` is considered to be a state sequence and will be displayed on top of the observations. Any column starting with the characters `state_prob` is considered to be the probability associated with the corresponding state at each time-point. That column will be used to alter the transparency of the state sequence visualization. Several state columns can be provided. For example: `"state_ground_truth"` and `"state_Viterbi"`. A state probability can be specified for each of the state column by providing, for example, the columns `"state_prob_ground_truth"` and `"state_prop_Viterbi"`.
model	a `hsmm` or `hsmm_spec` object specifying the model associated with the observation sequence.
title	(optional) a `character` specifying the title of the plot.
show_state_diff	(optional) a `logical` specifying if, in the case there are two "state" columns, a third line showing the agreement between these two columns should be displayed. Default value is `TRUE`. This is useful if one desires to compare a decoded sequence with the ground truth or state sequence resulting from the decoding of models with different parameters.
compact_view	(optional) a `logical` specifying if the visualization of the observed variables should be compact, i.e. using color-coding only to display each variable on a single line.
add_color_legend_in_compact_view	(optional) a `logical` specifying if the color legend should be added for each variable when displaying time-series in compact view. Default is `TRUE`.
add_state_color_legend	(optional) a `logical` specifying if the color legend for the model latent states should be printed. Default value is `FALSE`.
selection	(optional) a `data.frame` specifying the start and end of a "selection", i.e. a part of sequence that needs to be highlighted. If not `NULL` (the default value), this option allows to display a transparent rectangle across all state and variable lines of a color of a given state. The `data.frame` must have the following columns: `start, end, state (integer)`.
verbose	(optional) a logical specifying if the internal steps of the function should be printed.

Value

a ggplot object.

Examples

my_model = simple_model
Xsim = simulate_hsmm(model = my_model, n_state_transition = 20)
plot_hsmm_seq(X = Xsim, model = my_model)
plot_hsmm_seq(X = Xsim, model = my_model, title = "Simulated sequence", add_state_color_legend = TRUE)
plot_hsmm_seq(X = Xsim, model = my_model, title = "Simulated sequence (compact view)", compact_view = TRUE)