Merge pull request #334 from easystats/cor_sort_improvements

[Feature] cor_sort() can deal with non-square matrices

Author: strengejacke

DESCRIPTION

@@ -1,7 +1,7 @@
 Type: Package
 Package: correlation
 Title: Methods for Correlation Analysis
-Version: 0.8.6
+Version: 0.8.6.1
 Authors@R:
     c(person(given = "Dominique",
              family = "Makowski",
@@ -57,8 +57,8 @@ Imports:
     bayestestR (>= 0.15.0),
     datasets,
     datawizard (>= 0.13.0),
-    insight (>= 0.20.5),
-    parameters (>= 0.22.2),
+    insight (>= 1.0.0),
+    parameters (>= 0.24.0),
     stats
 Suggests:
     BayesFactor,

R/cor_sort.R

@@ -26,9 +26,9 @@ cor_sort <- function(x, distance = "correlation", hclust_method = "complete", ..
 
 #' @export
 cor_sort.easycorrelation <- function(x, distance = "correlation", hclust_method = "complete", ...) {
-  col_order <- .cor_sort_order(as.matrix(x), distance = distance, hclust_method = hclust_method, ...)
-  x$Parameter1 <- factor(x$Parameter1, levels = col_order)
-  x$Parameter2 <- factor(x$Parameter2, levels = col_order)
+  m <- cor_sort(as.matrix(x), distance = distance, hclust_method = hclust_method, ...)
+  x$Parameter1 <- factor(x$Parameter1, levels = rownames(m))
+  x$Parameter2 <- factor(x$Parameter2, levels = colnames(m))
   reordered <- x[order(x$Parameter1, x$Parameter2), ]
 
   # Restore class and attributes
@@ -38,6 +38,8 @@ cor_sort.easycorrelation <- function(x, distance = "correlation", hclust_method
   )
 
   # Make sure Parameter columns are character
+  # Was added to fix a test, but makes the function not work
+  # (See https://github.com/easystats/correlation/issues/259)
   # reordered$Parameter1 <- as.character(reordered$Parameter1)
   # reordered$Parameter2 <- as.character(reordered$Parameter2)
 
@@ -55,18 +57,32 @@ cor_sort.easycormatrix <- function(x, distance = "correlation", hclust_method =
   m <- x
   row.names(m) <- x$Parameter
   m <- as.matrix(m[names(m)[names(m) != "Parameter"]])
-  col_order <- .cor_sort_order(m, distance = distance, hclust_method = hclust_method, ...)
+
+  # If non-redundant matrix, fail (## TODO: fix that)
+  if (anyNA(m)) {
+    insight::format_error("Non-redundant matrices are not supported yet. Try again by setting summary(..., redundant = TRUE)")
+  }
+
+  # Get sorted matrix
+  m <- cor_sort(m, distance = distance, hclust_method = hclust_method, ...)
 
   # Reorder
-  x$Parameter <- factor(x$Parameter, levels = col_order)
-  reordered <- x[order(x$Parameter), c("Parameter", col_order)]
+  x$Parameter <- factor(x$Parameter, levels = row.names(m))
+  reordered <- x[order(x$Parameter), c("Parameter", colnames(m))]
 
   # Restore class and attributes
   attributes(reordered) <- utils::modifyList(
     attributes(x)[!names(attributes(x)) %in% c("names", "row.names")],
     attributes(reordered)
   )
 
+  # Reorder attributes (p-values) etc.
+  for (id in c("p", "CI", "CI_low", "CI_high", "BF", "Method", "n_Obs", "df_error", "t")) {
+    if (id %in% names(attributes(reordered))) {
+      attributes(reordered)[[id]] <- attributes(reordered)[[id]][order(x$Parameter), names(reordered)]
+    }
+  }
+
   # make sure Parameter columns are character
   reordered$Parameter <- as.character(reordered$Parameter)
 
@@ -76,8 +92,13 @@ cor_sort.easycormatrix <- function(x, distance = "correlation", hclust_method =
 
 #' @export
 cor_sort.matrix <- function(x, distance = "correlation", hclust_method = "complete", ...) {
-  col_order <- .cor_sort_order(x, distance = distance, hclust_method = hclust_method, ...)
-  reordered <- x[col_order, col_order]
+  if (isSquare(x) && all(colnames(x) %in% rownames(x))) {
+    i <- .cor_sort_square(x, distance = distance, hclust_method = hclust_method, ...)
+  } else {
+    i <- .cor_sort_nonsquare(x, distance = "euclidean", ...)
+  }
+
+  reordered <- x[i$row_order, i$col_order]
 
   # Restore class and attributes
   attributes(reordered) <- utils::modifyList(
@@ -91,7 +112,7 @@ cor_sort.matrix <- function(x, distance = "correlation", hclust_method = "comple
 # Utils -------------------------------------------------------------------
 
 
-.cor_sort_order <- function(m, distance = "correlation", hclust_method = "complete", ...) {
+.cor_sort_square <- function(m, distance = "correlation", hclust_method = "complete", ...) {
   if (distance == "correlation") {
     d <- stats::as.dist((1 - m) / 2) # r = -1 -> d = 1; r = 1 -> d = 0
   } else if (distance == "raw") {
@@ -101,5 +122,54 @@ cor_sort.matrix <- function(x, distance = "correlation", hclust_method = "comple
   }
 
   hc <- stats::hclust(d, method = hclust_method)
-  row.names(m)[hc$order]
+  row_order <- row.names(m)[hc$order]
+  list(row_order = row_order, col_order = row_order)
+}
+
+
+.cor_sort_nonsquare <- function(m, distance = "euclidean", ...) {
+  # Step 1: Perform clustering on rows and columns independently
+  row_dist <- stats::dist(m, method = distance) # Distance between rows
+  col_dist <- stats::dist(t(m), method = distance) # Distance between columns
+
+  row_hclust <- stats::hclust(row_dist, method = "average")
+  col_hclust <- stats::hclust(col_dist, method = "average")
+
+  # Obtain clustering orders
+  row_order <- row_hclust$order
+  col_order <- col_hclust$order
+
+  # Reorder matrix based on clustering
+  clustered_matrix <- m[row_order, col_order]
+
+  # Step 2: Refine alignment to emphasize strong correlations along the diagonal
+  n_rows <- nrow(clustered_matrix)
+  n_cols <- ncol(clustered_matrix)
+
+  used_rows <- logical(n_rows)
+  refined_row_order <- integer(0)
+
+  for (col in seq_len(n_cols)) {
+    max_value <- -Inf
+    best_row <- NA
+
+    for (row in seq_len(n_rows)[!used_rows]) {
+      if (abs(clustered_matrix[row, col]) > max_value) {
+        max_value <- abs(clustered_matrix[row, col])
+        best_row <- row
+      }
+    }
+
+    if (!is.na(best_row)) {
+      refined_row_order <- c(refined_row_order, best_row)
+      used_rows[best_row] <- TRUE
+    }
+  }
+
+  # Append any unused rows at the end
+  refined_row_order <- c(refined_row_order, which(!used_rows))
+
+  # Apply
+  m <- clustered_matrix[refined_row_order, ]
+  list(row_order = rownames(m), col_order = colnames(m))
 }

R/correlation.R

@@ -180,10 +180,9 @@
 #' `stats` package are supported.
 #' }
 #'
-#' @examplesIf requireNamespace("poorman", quietly = TRUE) && requireNamespace("psych", quietly = TRUE)
-#'
+#' @examplesIf all(insight::check_if_installed(c("psych", "datawizard"), quietly = TRUE)) && getRversion() >= "4.1.0"
 #' library(correlation)
-#' library(poorman)
+#' data(iris)
 #'
 #' results <- correlation(iris)
 #'
@@ -192,22 +191,20 @@
 #' summary(results, redundant = TRUE)
 #'
 #' # pipe-friendly usage with  grouped dataframes from {dplyr} package
-#' iris %>%
+#' iris |>
 #'   correlation(select = "Petal.Width", select2 = "Sepal.Length")
 #'
 #' # Grouped dataframe
 #' # grouped correlations
-#' iris %>%
-#'   group_by(Species) %>%
+#' iris |>
+#'   datawizard::data_group(Species) |>
 #'   correlation()
 #'
 #' # selecting specific variables for correlation
-#' mtcars %>%
-#'   group_by(am) %>%
-#'   correlation(
-#'     select = c("cyl", "wt"),
-#'     select2 = c("hp")
-#'   )
+#' data(mtcars)
+#' mtcars |>
+#'   datawizard::data_group(am) |>
+#'   correlation(select = c("cyl", "wt"), select2 = "hp")
 #'
 #' # supplying custom variable names
 #' correlation(anscombe, select = c("x1", "x2"), rename = c("var1", "var2"))
@@ -425,8 +422,36 @@ correlation <- function(data,
   ungrouped_x <- as.data.frame(data)
   xlist <- split(ungrouped_x, ungrouped_x[groups], sep = " - ")
 
-  # If data 2 is provided
-  if (!is.null(data2)) {
+  # If data 2 is not provided
+  if (is.null(data2)) {
+    modelframe <- data.frame()
+    out <- data.frame()
+    for (i in names(xlist)) {
+      xlist[[i]][groups] <- NULL
+      rez <- .correlation(
+        xlist[[i]],
+        data2,
+        method = method,
+        p_adjust = p_adjust,
+        ci = ci,
+        bayesian = bayesian,
+        bayesian_prior = bayesian_prior,
+        bayesian_ci_method = bayesian_ci_method,
+        bayesian_test = bayesian_test,
+        redundant = redundant,
+        include_factors = include_factors,
+        partial = partial,
+        partial_bayesian = partial_bayesian,
+        multilevel = multilevel,
+        ranktransform = ranktransform,
+        winsorize = winsorize
+      )
+      modelframe_current <- rez$data
+      rez$params$Group <- modelframe_current$Group <- i
+      out <- rbind(out, rez$params)
+      modelframe <- rbind(modelframe, modelframe_current)
+    }
+  } else {
     if (inherits(data2, "grouped_df")) {
       groups2 <- setdiff(colnames(attributes(data2)$groups), ".rows")
       if (!all.equal(groups, groups2)) {
@@ -463,35 +488,6 @@ correlation <- function(data,
         modelframe <- rbind(modelframe, modelframe_current)
       }
     }
-    # else
-  } else {
-    modelframe <- data.frame()
-    out <- data.frame()
-    for (i in names(xlist)) {
-      xlist[[i]][groups] <- NULL
-      rez <- .correlation(
-        xlist[[i]],
-        data2,
-        method = method,
-        p_adjust = p_adjust,
-        ci = ci,
-        bayesian = bayesian,
-        bayesian_prior = bayesian_prior,
-        bayesian_ci_method = bayesian_ci_method,
-        bayesian_test = bayesian_test,
-        redundant = redundant,
-        include_factors = include_factors,
-        partial = partial,
-        partial_bayesian = partial_bayesian,
-        multilevel = multilevel,
-        ranktransform = ranktransform,
-        winsorize = winsorize
-      )
-      modelframe_current <- rez$data
-      rez$params$Group <- modelframe_current$Group <- i
-      out <- rbind(out, rez$params)
-      modelframe <- rbind(modelframe, modelframe_current)
-    }
   }
 
   # Group as first column

R/display.R

@@ -2,7 +2,8 @@
 #' @name display.easycormatrix
 #'
 #' @description Export tables (i.e. data frame) into different output formats.
-#'   `print_md()` is a alias for `display(format = "markdown")`.
+#'   `print_md()` is a alias for `display(format = "markdown")`. Note that
+#'   you can use `format()` to get the formatted table as a dataframe.
 #'
 #' @param object,x An object returned by
 #'   [`correlation()`][correlation] or its summary.

R/methods_format.R

@@ -156,10 +156,17 @@ format.easycormatrix <- function(x,
   # final new line
   footer <- paste0(footer, "\n")
 
-  # for html/markdown, create list
+  # for html/markdown, modify footer format
   if (!is.null(format) && format != "text") {
+    # no line break if not text format
     footer <- unlist(strsplit(footer, "\n", fixed = TRUE))
-    footer <- as.list(footer[nzchar(footer, keepNA = TRUE)])
+    # remove empty elements
+    footer <- footer[nzchar(footer, keepNA = TRUE)]
+    # create list or separate by ";"
+    footer <- switch(format,
+      html = paste(footer, collapse = "; "),
+      as.list(footer)
+    )
   }
 
   footer
@@ -168,7 +175,9 @@ format.easycormatrix <- function(x,
 
 #' @keywords internal
 .format_easycorrelation_caption <- function(x, format = NULL) {
-  if (!is.null(attributes(x)$method)) {
+  if (is.null(attributes(x)$method)) {
+    caption <- NULL
+  } else {
     if (isTRUE(attributes(x)$smoothed)) {
       prefix <- "Smoothed Correlation Matrix ("
     } else {
@@ -179,8 +188,6 @@ format.easycormatrix <- function(x,
     } else {
       caption <- paste0(prefix, unique(attributes(x)$method), "-method)")
     }
-  } else {
-    caption <- NULL
   }
 
   caption

R/methods_print.R

@@ -3,7 +3,7 @@
 
 #' @export
 print.easycorrelation <- function(x, ...) {
-  cat(insight::export_table(format(x, ...), format = "text"))
+  cat(insight::export_table(format(x, ...), ...))
   invisible(x)
 }
 
@@ -13,9 +13,9 @@ print.easycormatrix <- function(x, ...) {
   # If real matrix, print as matrix
   if (colnames(formatted)[1] == "Variables") {
     formatted$Variables <- NULL
-    print(as.matrix(formatted))
+    print(as.matrix(formatted), ...)
   } else {
-    cat(insight::export_table(format(x, ...), format = "text"))
+    cat(insight::export_table(format(x, ...), ...))
   }
   invisible(x)
 }
@@ -31,7 +31,7 @@ print.easymatrixlist <- function(x, cols = "auto", ...) {
 
   for (i in cols) {
     cat(" ", i, " ", "\n", rep("-", nchar(i) + 2), "\n", sep = "")
-    print(x[[i]])
+    print(x[[i]], ...)
     cat("\n")
   }
 }
@@ -40,7 +40,7 @@ print.easymatrixlist <- function(x, cols = "auto", ...) {
 print.grouped_easymatrixlist <- function(x, cols = "auto", ...) {
   for (i in names(x)) {
     cat(rep("=", nchar(i) + 2), "\n ", i, " ", "\n", rep("=", nchar(i) + 2), "\n\n", sep = "")
-    print(x[[i]])
+    print(x[[i]], ...)
     cat("\n")
   }
 }

correlation.Rproj

@@ -1,4 +1,5 @@
 Version: 1.0
+ProjectId: a2737226-16da-4377-8659-b462bf604f1e
 
 RestoreWorkspace: No
 SaveWorkspace: No