Merge 4772a6d into 005d4c7

CHANGELOG.rst

@@ -1,6 +1,7 @@
 In development
 --------------
-
+- Added option to display both the absolute frequency and the percentage of
+  the total for each intersection and category.
 - Improved efficiency where there are many categories, but valid combinations
   are sparse, if `sort_by='degree'`. (:issue:`82`)
 - Permit truthy (not necessarily bool) values in index. (:issue:`74`)

upsetplot/__init__.py

@@ -1,4 +1,4 @@
-__version__ = '0.4-dev'
+__version__ = '0.4.dev1'
 
 import os
 

upsetplot/plotting.py

@@ -272,6 +272,8 @@ class UpSet:
         Whether to label the intersection size bars with the cardinality
         of the intersection. When a string, this formats the number.
         For example, '%d' is equivalent to True.
+        '%' or 'pctg' or 'percentage' displays both the absolute frequency and
+        the percentage of the total for each intersection and category
     sort_sets_by
         .. deprecated: 0.3
             Replaced by sort_categories_by, this parameter will be removed in
@@ -285,7 +287,8 @@ def __init__(self, data, orientation='horizontal', sort_by='degree',
                  facecolor='black',
                  with_lines=True, element_size=32,
                  intersection_plot_elements=6, totals_plot_elements=2,
-                 show_counts='', sort_sets_by='deprecated'):
+                 show_counts='', show_percentages=False,
+                 sort_sets_by='deprecated'):
 
         self._horizontal = orientation == 'horizontal'
         self._reorient = _identity if self._horizontal else _transpose
@@ -299,6 +302,7 @@ def __init__(self, data, orientation='horizontal', sort_by='degree',
         if not intersection_plot_elements:
             self._subset_plots.pop()
         self._show_counts = show_counts
+        self._show_percentages = show_percentages
 
         if sort_sets_by != 'deprecated':
             sort_categories_by = sort_sets_by
@@ -429,7 +433,7 @@ def make_grid(self, fig=None):
                                       self._totals_plot_elements),
                       hspace=1)
         if self._horizontal:
-            print(n_cats, n_inters, self._totals_plot_elements)
+            # print(n_cats, n_inters, self._totals_plot_elements)
             out = {'matrix': gridspec[-n_cats:, -n_inters:],
                    'shading': gridspec[-n_cats:, :],
                    'totals': gridspec[-n_cats:, :self._totals_plot_elements],
@@ -446,8 +450,8 @@ def make_grid(self, fig=None):
             cumsizes = np.cumsum(sizes)
             for start, stop, plot in zip(np.hstack([[0], cumsizes]), cumsizes,
                                          self._subset_plots):
-                out[plot['id']] = gridspec[-n_inters:,
-                                           start + n_cats:stop + n_cats]
+                out[plot['id']] = \
+                    gridspec[-n_inters:, start + n_cats:stop + n_cats]
         return out
 
     def plot_matrix(self, ax):
@@ -505,32 +509,66 @@ def plot_intersections(self, ax):
         ax.set_ylabel('Intersection size')
 
     def _label_sizes(self, ax, rects, where):
-        if not self._show_counts:
+        if not self._show_counts and not self._show_percentages:
             return
         fmt = '%d' if self._show_counts is True else self._show_counts
         if where == 'right':
             margin = 0.01 * abs(np.diff(ax.get_xlim()))
             for rect in rects:
                 width = rect.get_width()
-                ax.text(width + margin,
-                        rect.get_y() + rect.get_height() * .5,
-                        fmt % width,
-                        ha='left', va='center')
+                if self._show_percentages:
+                    ax.text(width + margin,
+                            rect.get_y() + rect.get_height() * .5,
+                            fmt % width +
+                            " (%.1f%%)" % (100 * width /
+                                           max(1, sum(self.intersections)))
+                            if self._show_counts else
+                            "%.1f%%" % (100 * width /
+                                        max(1, sum(self.intersections))),
+                            ha='left', va='center')
+                else:
+                    ax.text(width + margin,
+                            rect.get_y() + rect.get_height() * .5,
+                            fmt % width,
+                            ha='left', va='center')
         elif where == 'left':
             margin = 0.01 * abs(np.diff(ax.get_xlim()))
             for rect in rects:
                 width = rect.get_width()
-                ax.text(width + margin,
-                        rect.get_y() + rect.get_height() * .5,
-                        fmt % width,
-                        ha='right', va='center')
+                if self._show_percentages:
+                    ax.text(width + margin,
+                            rect.get_y() + rect.get_height() * .5,
+                            fmt % width +
+                            " (%.1f%%)" % (100 * width /
+                                           max(1, sum(self.intersections)))
+                            if self._show_counts
+                            else "%.1f%%" % (100 * width /
+                                             max(1, sum(self.intersections))),
+                            ha='right', va='center')
+                else:
+                    ax.text(width + margin,
+                            rect.get_y() + rect.get_height() * .5,
+                            fmt % width,
+                            ha='right', va='center')
         elif where == 'top':
             margin = 0.01 * abs(np.diff(ax.get_ylim()))
             for rect in rects:
                 height = rect.get_height()
-                ax.text(rect.get_x() + rect.get_width() * .5,
-                        height + margin, fmt % height,
-                        ha='center', va='bottom')
+                if self._show_percentages:
+                    ax.text(rect.get_x() + rect.get_width() * .5,
+                            height + margin,
+                            fmt % height +
+                            "\n(%.1f%%)" % (100 * height /
+                                            max(1, sum(self.intersections)))
+                            if self._show_counts
+                            else "%.1f%%" % (100 * height /
+                                             max(1, sum(self.intersections))),
+                            ha='center', va='bottom', fontsize=8)
+                else:
+                    ax.text(rect.get_x() + rect.get_width() * .5,
+                            height + margin,
+                            fmt % height,
+                            ha='center', va='bottom')
         else:
             raise NotImplementedError('unhandled where: %r' % where)