From 716d47d46fa708a35f52c9681b39bd6b643df9a6 Mon Sep 17 00:00:00 2001
From: Matt Dancho <mdancho@gmail.com>
Date: Sat, 28 Oct 2023 20:06:41 -0400
Subject: [PATCH] plot_anomalies_decomp

---
 R/anomalize-plotting.R | 15 +++++-----
 R/anomalize.R          | 62 ++++++++++++++++++++++++++++++++----------
 _pkgdown.yml           |  2 +-
 man/anomalize.Rd       |  8 +++---
 man/plot_anomalies.Rd  | 12 ++++----
 5 files changed, 65 insertions(+), 34 deletions(-)

diff --git a/R/anomalize-plotting.R b/R/anomalize-plotting.R
index c8085ad7..dff40e5a 100644
--- a/R/anomalize-plotting.R
+++ b/R/anomalize-plotting.R
@@ -2,7 +2,7 @@
 
 #' Visualize Anomalies for One or More Time Series
 #'
-#' An interactive and scalable function for visualizing anomalies in time series data.
+#' `plot_anomalies()` is an interactive and scalable function for visualizing anomalies in time series data.
 #' Plots are available in interactive `plotly` (default) and static `ggplot2` format.
 #'
 #' @param .data A `tibble` or `data.frame` that has been anomalized by `anomalize()`
@@ -47,6 +47,7 @@
 #'
 #'
 #' @examples
+#' # Plot Anomaly
 #' library(dplyr)
 #'
 #' walmart_sales_weekly %>%
@@ -360,8 +361,8 @@ plot_anomalies.grouped_df <- function(
 # 2.0 PLOT ANOMALIES DECOMP ----
 #' Visualize Anomaly Decomposition
 #'
-#' An interactive and scalable function for visualizing anomaly decomposition in time series data.
-#' Plots are available in interactive `plotly` (default) and static `ggplot2` format.
+#' `plot_anomalies_decomp` function takes in data from the `anomalize()`
+#' function, and returns a plot of the anomaly decomposition.
 #'
 #' @param .data A `tibble` or `data.frame` that has been anomalized by `anomalize()`
 #' @param .date_var A column containing either date or date-time values
@@ -400,11 +401,8 @@ plot_anomalies.grouped_df <- function(
 #'  - `scales`: use `facet_scales`
 #'  - `as_plotly`: use `.interactive`
 #'
-#'
-#' @return A `plotly` or `ggplot2` visualization
-#'
-#'
 #' @examples
+#' # Plot Anomaly Decomposition
 #' library(dplyr)
 #'
 #' walmart_sales_weekly %>%
@@ -610,6 +608,7 @@ plot_anomalies_decomp.grouped_df <- function(
 
     return(g)
 
-
 }
 
+
+
diff --git a/R/anomalize.R b/R/anomalize.R
index 785e15d0..9bb085b9 100644
--- a/R/anomalize.R
+++ b/R/anomalize.R
@@ -101,11 +101,22 @@
 #' @name anomalize
 #' @export
 #'
-anomalize <- function(.data, .date_var, .value,
-                      .frequency = "auto", .trend = "auto",
-                      .method = "stl",
-                      .iqr_alpha = 0.05, .clean_alpha = 0.75, .max_anomalies = 0.2,
-                      .message = TRUE) {
+anomalize <- function(
+        .data,
+        .date_var,
+        .value,
+
+        .frequency = "auto",
+        .trend = "auto",
+
+        method = "stl",
+
+        .iqr_alpha = 0.05,
+        .clean_alpha = 0.75,
+        .max_anomalies = 0.2,
+
+        .message = TRUE
+) {
 
     # Tidyeval Setup
     date_var_expr <- rlang::enquo(.date_var)
@@ -132,11 +143,22 @@ anomalize <- function(.data, .date_var, .value,
 
 
 #' @export
-anomalize.data.frame <- function(.data, .date_var, .value,
-                                 .frequency = "auto", .trend = "auto",
-                                 .method = "stl",
-                                 .iqr_alpha = 0.05, .clean_alpha = 0.75, .max_anomalies = 0.2,
-                                 .message = TRUE) {
+anomalize.data.frame <- function(
+    .data,
+    .date_var,
+    .value,
+
+    .frequency = "auto",
+    .trend = "auto",
+
+    method = "stl",
+
+    .iqr_alpha = 0.05,
+    .clean_alpha = 0.75,
+    .max_anomalies = 0.2,
+
+    .message = TRUE
+) {
 
     # STL Decomposition (observed, season, trend, remainder, seasadj)
     ret <- .data %>%
@@ -181,12 +203,22 @@ anomalize.data.frame <- function(.data, .date_var, .value,
 }
 
 #' @export
-anomalize.grouped_df <- function(.data, .date_var, .value,
-                                 .frequency = "auto", .trend = "auto",
-                                 .method = "stl",
-                                 .iqr_alpha = 0.05, .clean_alpha = 0.75, .max_anomalies = 0.2,
-                                 .message = TRUE) {
+anomalize.grouped_df <- function(
+        .data,
+        .date_var,
+        .value,
+
+        .frequency = "auto",
+        .trend = "auto",
+
+        method = "stl",
+
+        .iqr_alpha = 0.05,
+        .clean_alpha = 0.75,
+        .max_anomalies = 0.2,
 
+        .message = TRUE
+) {
 
     # Tidy Eval Setup
     value_expr    <- rlang::enquo(.value)
diff --git a/_pkgdown.yml b/_pkgdown.yml
index b7787676..73ef1942 100644
--- a/_pkgdown.yml
+++ b/_pkgdown.yml
@@ -114,7 +114,7 @@ reference:
 - subtitle: Anomaly Detection
   contents:
   - anomalize
-  - plot_anomalies
+  - contains("plot_anomalies")
 - subtitle: Function Operations
   contents: slidify
 - subtitle: Vector Operations
diff --git a/man/anomalize.Rd b/man/anomalize.Rd
index 9daca6fc..1e2f7943 100644
--- a/man/anomalize.Rd
+++ b/man/anomalize.Rd
@@ -10,7 +10,7 @@ anomalize(
   .value,
   .frequency = "auto",
   .trend = "auto",
-  .method = "stl",
+  method = "stl",
   .iqr_alpha = 0.05,
   .clean_alpha = 0.75,
   .max_anomalies = 0.2,
@@ -33,9 +33,6 @@ Refer to \code{\link[=tk_get_frequency]{tk_get_frequency()}}.}
 For STL, trend controls the sensitivity of the LOESS smoother, which is used to remove the remainder.
 Refer to \code{\link[=tk_get_trend]{tk_get_trend()}}.}
 
-\item{.method}{The outlier detection method. Default: "stl". Currently
-"stl" is the only method. "twitter" is planned.}
-
 \item{.iqr_alpha}{Controls the width of the "normal" range. Lower values are more conservative
 while higher values are less prone to incorrectly classifying "normal" observations.}
 
@@ -48,6 +45,9 @@ depending on the direction of the anomaly.}
 
 \item{.message}{A boolean. If \code{TRUE}, will output information related to automatic frequency
 and trend selection (if applicable).}
+
+\item{.method}{The outlier detection method. Default: "stl". Currently
+"stl" is the only method. "twitter" is planned.}
 }
 \value{
 A \code{tibble} or \code{data.frame} with the following columns:
diff --git a/man/plot_anomalies.Rd b/man/plot_anomalies.Rd
index 1871ef1f..8c66de14 100644
--- a/man/plot_anomalies.Rd
+++ b/man/plot_anomalies.Rd
@@ -117,18 +117,17 @@ The only parameters that cannot be passed are:
 }}
 }
 \value{
-A \code{plotly} or \code{ggplot2} visualization
-
 A \code{plotly} or \code{ggplot2} visualization
 }
 \description{
-An interactive and scalable function for visualizing anomalies in time series data.
+\code{plot_anomalies()} is an interactive and scalable function for visualizing anomalies in time series data.
 Plots are available in interactive \code{plotly} (default) and static \code{ggplot2} format.
 
-An interactive and scalable function for visualizing anomaly decomposition in time series data.
-Plots are available in interactive \code{plotly} (default) and static \code{ggplot2} format.
+\code{plot_anomalies_decomp} function takes in data from the \code{anomalize()}
+function, and returns a plot of the anomaly decomposition.
 }
 \examples{
+# Plot Anomaly
 library(dplyr)
 
 walmart_sales_weekly \%>\%
@@ -137,12 +136,13 @@ walmart_sales_weekly \%>\%
     anomalize(Date, Weekly_Sales) \%>\%
     plot_anomalies(Date, .facet_ncol = 2, .ribbon_alpha = 0.25, .interactive = FALSE)
 
+# Plot Anomaly Decomposition
 library(dplyr)
 
 walmart_sales_weekly \%>\%
     filter(id \%in\% c("1_1", "1_3")) \%>\%
     group_by(id) \%>\%
-    anomalize(Date, Weekly_Sales) \%>\%
+    anomalize(Date, Weekly_Sales, .message = FALSE) \%>\%
     plot_anomalies_decomp(Date, .interactive = FALSE)
 
 }