## ----setup, include = FALSE-----------------------------------------
knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>"
)
old_options <- options(width = 70)

## ----load-----------------------------------------------------------
library(deriva)

## ----quickstart-----------------------------------------------------
# Simulate a stream: 500 stable observations, then 500 with higher error rate
stream <- sim_drift_stream(n_pre = 500, n_post = 500,
                           p_pre = 0.05, p_post = 0.30,
                           seed = 42)

result <- detect_drift(stream, .col = error, method = "ddm")

# Where was drift flagged?
subset(result, .drift)

## ----spec-----------------------------------------------------------
spec <- drift_detector("ddm", min_instances = 30)
spec

## ----fit------------------------------------------------------------
baseline <- sim_drift_stream(n_pre = 300, n_post = 0, seed = 1)

fitted <- fit(spec, baseline, signal = error)
fitted

## ----advance--------------------------------------------------------
batch1 <- sim_drift_stream(n_pre = 200, n_post = 0,   seed = 2)
batch2 <- sim_drift_stream(n_pre = 0,   n_post = 300, p_post = 0.35, seed = 3)

fitted2 <- advance(fitted,  batch1)
fitted3 <- advance(fitted2, batch2)
fitted3

## ----augment--------------------------------------------------------
history <- augment(fitted3)
tail(history[, c("t", "error", ".phase", ".warning", ".drift")], 10)

## ----tidy-----------------------------------------------------------
tidy(fitted3)

## ----glance---------------------------------------------------------
glance(fitted3)

## ----autoplot, eval = requireNamespace("ggplot2", quietly = TRUE), fig.width = 6.5, fig.height = 3.5----
library(ggplot2)
autoplot(fitted3)

## ----bridge---------------------------------------------------------
# Simulate tidymodels augment() output for a classifier
predictions <- data.frame(
  time     = 1:8,
  truth    = factor(c("yes","no","yes","yes","no","yes","no","yes")),
  .pred_class = factor(c("yes","no","yes","no" ,"no","no" ,"no","yes"))
)

add_prediction_error(predictions, truth = truth)

## ----kswin----------------------------------------------------------
cont_stream <- sim_dist_stream(
  n_pre = 500, n_post = 500,
  mean_pre = 0, mean_post = 2,
  seed = 99
)

detect_drift(cont_stream, .col = value, method = "kswin") |>
  subset(.drift) |>
  head()

## ----cleanup, include = FALSE-------------------------------------------------
options(old_options)