The high resolution of bio-logging sensor data relative to the duration of deployments limits the utility of static figures. sb_plot_data and sb_plot_predictions use plotly to create interactive figures for looking in-depth at bio-logging data.

sb_plot_data(deployid, sensors, events)

sb_plot_predictions(deployid, sensors, predictions, outcomes = NULL)

Arguments

deployid

[character(1)] ID of the deployment to visualize. Must be a deployment in both sensors and events.

sensors

[Sensors] Bio-logging sensor data (see Sensors).

events

[Events] Labeled behavioral events (see Events).

predictions

[Predictions] Predicted behavioral events (see sb_predict).

outcomes

[Outcomes] Outcomes of predicted behavioral events (see sb_assess) (optional).

Value

An interactive figure with time on the x-axis and a separate y-axis for each bio-logging sensor variable, arranged vertically. Interaction options include: click-and-drag to zoom in, double-click to zoom out, hover to see details (exact date, time, and value).

sb_plot_data

Points indicate known behavioral events.

sb_plot_predictions

Solid points indicate predicted behavioral events. If actual events are known (i.e, outcomes is not NULL), they are plotted as hollow circles and color-coded as follows: true positive as blue solid point in blue hollow point, false positive as red solid point, false negative as red hollow point. In addition to the bio-logging sensor variables, there's an additional time series for the "local probability" of an event (see vignette(TODO)).

Examples

# Load sample data and split test/train
c(lunge_sensors, lunge_events) %<-% load_lunges()
test_deployids <- deployments(lunge_sensors)[1:3]
deployid <- test_deployids[1]
c(sensors_test, sensors_train) %<-% divide(lunge_sensors, test_deployids)
c(events_test, events_train) %<-% divide(lunge_events, test_deployids)

# Visualize sensor and event data for one deployment
sb_plot_data(deployid, lunge_sensors, lunge_events)