Sankey Diagram and bca warning

Finalized Sankey diagram plotting and add quantile statistics warning and confidence interval plotting pararmeter

Author: Jacobluke-

dabest/_classes.py

@@ -1570,7 +1570,8 @@ class TwoGroupsEffectSize(object):
     mean differences between two groups.
     """
 
-    def __init__(self, control, test, effect_size,proportional,
+    def __init__(self, control, test, effect_size,
+                 proportional=False,
                  is_paired=None, ci=95,
                  resamples=5000, 
                  permutation_count=5000, 
@@ -2662,7 +2663,7 @@ def plot(self, color_col=None,
             #bar plot
             bar_label=None, bar_desat=0.5, bar_width = 0.5,bar_ylim = None,
             # error bar of proportion plot
-            ci=None, err_color=None,
+            ci=None, ci_type='bca', err_color=None,
 
             float_contrast=True,
             show_pairs=True,

dabest/_stats_tools/effsize.py

@@ -7,6 +7,7 @@
 
     two_group_difference
     cohens_d
+    cohens_h
     hedges_g
     cliffs_delta
     func_difference
@@ -71,11 +72,19 @@ def two_group_difference(control, test, is_paired=False,
         float: The desired effect size.
     """
     import numpy as np
+    import warnings
 
     if effect_size == "mean_diff":
         return func_difference(control, test, np.mean, is_paired)
 
     elif effect_size == "median_diff":
+        mes1 = "Using median as the statistic in bootstrapping may \
+                result in a biased estimate and cause problems with \
+                BCa confidence intervals. Consider using a different statistic, such as the mean.\n"
+        mes2 = "When plotting, please consider using percetile confidence intervals\
+                by specifying `ci_type='percentile'`. For detailed information, \
+                refer to https://github.com/ACCLAB/DABEST-python/issues/129"
+        warnings.warn(message=mes1+mes2, category=UserWarning)
         return func_difference(control, test, np.median, is_paired)
 
     elif effect_size == "cohens_d":
@@ -257,7 +266,7 @@ def cohens_h(control, test):
     import pandas as pd
 
     # Check whether dataframe contains only 0s and 1s.
-    if pd.unique(control)==np.array([0,1]).all()==False and (pd.unique(test)==np.array([0,1])).all()==False:
+    if np.isin(control, [0, 1]).all() == False or np.isin(test, [0, 1]).all() == False:
         raise ValueError("Input data must be binary.")
 
     # Convert to numpy arrays for speed.

dabest/plot_tools.py

@@ -449,6 +449,9 @@ def single_sankey(left, right, xpos=0, leftWeight=None, rightWeight=None,
         vertical extent of the diagram in units of horizontal extent
     rightColor: bool
         if True, each strip of the diagram will be colored according to the corresponding left labels
+    align: bool
+        if 'center', the diagram will be centered on each xtick, 
+        if 'edge', the diagram will be aligned with the left edge of each xtick
     '''
 
     from collections import defaultdict
@@ -510,6 +513,8 @@ def single_sankey(left, right, xpos=0, leftWeight=None, rightWeight=None,
         colorPalette = sns.color_palette(palette, len(allLabels))
         for i, label in enumerate(allLabels):
             colorDict[label] = colorPalette[i]
+        fail_color = {0:"grey"}
+        colorDict.update(fail_color)
     else:
         missing = [label for label in allLabels if label not in colorDict.keys()]
         if missing:
@@ -527,6 +532,8 @@ def single_sankey(left, right, xpos=0, leftWeight=None, rightWeight=None,
             raise TypeError(f'the dtypes of parameters x ({xpos.dtype}) '
                             f'and width ({width.dtype}) '
                             f'are incompatible') from e
+    else: 
+        leftpos = xpos
 
 
     # Determine positions of left label patches and total widths
@@ -640,13 +647,14 @@ def normalize_dict(nested_dict, target):
                 rightWidths_norm[rightLabel]['bottom'] += ns_r_norm[leftLabel][rightLabel]
                 ax.fill_between(
                     np.linspace(leftpos, leftpos + xMax, len(ys_d)), ys_d, ys_u, alpha=alpha,
-                    color=colorDict[labelColor]
+                    color=colorDict[labelColor], edgecolor='none'
                 )
 
 def sankeydiag(data, xvar, yvar, left_idx, right_idx, 
                 leftLabels=None, rightLabels=None,  
-                palette=None,
-                ax=None, width=0.5, rightColor=False,
+                palette=None, ax=None, 
+                one_sankey=False,
+                width=0.5, rightColor=False,
                 align='center', alpha=0.65, **kwargs):
     '''
     Read in melted pd.DataFrame, and draw multiple sankey diagram on a single axes
@@ -670,6 +678,9 @@ def sankeydiag(data, xvar, yvar, left_idx, right_idx,
         labels for the right side of the diagram. The diagram will be sorted by these labels.
     palette: str or dict
     ax: matplotlib axes to be drawn on
+    one_sankey: bool 
+        determined by the driver function on plotter.py. 
+        if True, draw the sankey diagram across the whole raw data axes
     width: float
         the width of each sankey diagram
     align: str
@@ -703,7 +714,7 @@ def sankeydiag(data, xvar, yvar, left_idx, right_idx,
         bar_width = kwargs["bar_width"]
 
     if ax is None:
-        fig, ax = plt.subplots()
+        ax = plt.gca()
 
     allLabels = pd.Series(np.sort(data[yvar].unique())[::-1]).unique()
 
@@ -740,13 +751,27 @@ def sankeydiag(data, xvar, yvar, left_idx, right_idx,
         plot_palette = None
 
     for left, right in zip(broadcasted_left, right_idx):
-        single_sankey(data[data[xvar]==left][yvar], data[data[xvar]==right][yvar], 
-                        xpos=xpos, ax=ax, colorDict=plot_palette, width=width, 
-                        leftLabels=leftLabels, rightLabels=rightLabels, 
-                        rightColor=rightColor, bar_width=bar_width,
-                        align=align, alpha=alpha)
-        xpos += 1
-
-    sankey_ticks = [f"{left}\n v.s.\n{right}" for left, right in zip(broadcasted_left, right_idx)]
-    ax.get_xaxis().set_ticks(np.arange(len(right_idx)))
-    ax.get_xaxis().set_ticklabels(sankey_ticks)
+        if one_sankey == False:
+            single_sankey(data[data[xvar]==left][yvar], data[data[xvar]==right][yvar], 
+                            xpos=xpos, ax=ax, colorDict=plot_palette, width=width, 
+                            leftLabels=leftLabels, rightLabels=rightLabels, 
+                            rightColor=rightColor, bar_width=bar_width,
+                            align=align, alpha=alpha)
+            xpos += 1
+        else:
+            xpos = 0 + bar_width/2
+            width = 1 - bar_width
+            single_sankey(data[data[xvar]==left][yvar], data[data[xvar]==right][yvar], 
+                            xpos=xpos, ax=ax, colorDict=plot_palette, width=width, 
+                            leftLabels=leftLabels, rightLabels=rightLabels, 
+                            rightColor=rightColor, bar_width=bar_width,
+                            align='edge', alpha=alpha)
+
+    if one_sankey == False:
+        sankey_ticks = [f"{left}\n v.s.\n{right}" for left, right in zip(broadcasted_left, right_idx)]
+        ax.get_xaxis().set_ticks(np.arange(len(right_idx)))
+        ax.get_xaxis().set_ticklabels(sankey_ticks)
+    else:
+        sankey_ticks = [broadcasted_left[0], right_idx[0]]
+        ax.set_xticks([0, 1])
+        ax.set_xticklabels(sankey_ticks)