AutoStatLib 0.2.13__tar.gz → 0.2.16__tar.gz

{autostatlib-0.2.13/src/AutoStatLib.egg-info → autostatlib-0.2.16}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: AutoStatLib
-Version: 0.2.13
+Version: 0.2.16
 Summary: AutoStatLib - a simple statistical analysis tool
 Author: Stemonitis, SciWare LLC
 Author-email: konung-yaropolk <yaropolk1995@gmail.com>
@@ -533,6 +533,7 @@ Requires-Dist: numpy
 Requires-Dist: scipy
 Requires-Dist: statsmodels
 Requires-Dist: matplotlib
+Requires-Dist: seaborn
 Requires-Dist: scikit-posthocs
 Requires-Dist: pandas
 Dynamic: license-file
@@ -675,22 +676,22 @@ If errors occured, *GetResult()* returns an empty dictionary
 -- add QQ plot
 -- n-sample tests: add onetail option
 
-✅ done -- detailed normality test results
-✅ done -- added posthoc: Kruskal-Wallis Dunn's multiple comparisons
-
-tests check:
-1-sample:
---Wilcoxon 2,1 tails - ✅ok
---t-tests 2,1 tails -✅ok
-
-2-sample:
---Wilcoxon 2,1 tails - ✅ok
---Mann-whitney 2,1 tails - ✅ok
---t-tests 2,1 tails -✅ok
-
-n-sample:
---Kruskal-Wallis 2 tail - ✅ok
---Dunn's multiple comparisons - ✅ok
---Friedman 2 tail - ✅ok
---one-way ANOWA 2 tail - ✅ok
---Tukey`s multiple comparisons - ✅ok
+✅ done -- detailed normality test results  
+✅ done -- added posthoc: Kruskal-Wallis Dunn's multiple comparisons  
+
+tests check:  
+1-sample:  
+--Wilcoxon 2,1 tails - ✅ok  
+--t-tests 2,1 tails -✅ok  
+
+2-sample:  
+--Wilcoxon 2,1 tails - ✅ok  
+--Mann-whitney 2,1 tails - ✅ok  
+--t-tests 2,1 tails -✅ok  
+
+n-sample:  
+--Kruskal-Wallis 2 tail - ✅ok  
+--Dunn's multiple comparisons - ✅ok   
+--Friedman 2 tail - ✅ok  
+--one-way ANOWA 2 tail - ✅ok  
+--Tukey`s multiple comparisons - ✅ok  

{autostatlib-0.2.13 → autostatlib-0.2.16}/README.md

@@ -136,22 +136,22 @@ If errors occured, *GetResult()* returns an empty dictionary
 -- add QQ plot
 -- n-sample tests: add onetail option
 
-✅ done -- detailed normality test results
-✅ done -- added posthoc: Kruskal-Wallis Dunn's multiple comparisons
-
-tests check:
-1-sample:
---Wilcoxon 2,1 tails - ✅ok
---t-tests 2,1 tails -✅ok
-
-2-sample:
---Wilcoxon 2,1 tails - ✅ok
---Mann-whitney 2,1 tails - ✅ok
---t-tests 2,1 tails -✅ok
-
-n-sample:
---Kruskal-Wallis 2 tail - ✅ok
---Dunn's multiple comparisons - ✅ok
---Friedman 2 tail - ✅ok
---one-way ANOWA 2 tail - ✅ok
---Tukey`s multiple comparisons - ✅ok
+✅ done -- detailed normality test results  
+✅ done -- added posthoc: Kruskal-Wallis Dunn's multiple comparisons  
+
+tests check:  
+1-sample:  
+--Wilcoxon 2,1 tails - ✅ok  
+--t-tests 2,1 tails -✅ok  
+
+2-sample:  
+--Wilcoxon 2,1 tails - ✅ok  
+--Mann-whitney 2,1 tails - ✅ok  
+--t-tests 2,1 tails -✅ok  
+
+n-sample:  
+--Kruskal-Wallis 2 tail - ✅ok  
+--Dunn's multiple comparisons - ✅ok   
+--Friedman 2 tail - ✅ok  
+--one-way ANOWA 2 tail - ✅ok  
+--Tukey`s multiple comparisons - ✅ok  

{autostatlib-0.2.13 → autostatlib-0.2.16}/requirements.txt

@@ -2,5 +2,6 @@ numpy
 scipy
 statsmodels
 matplotlib
+seaborn
 scikit-posthocs
 pandas

{autostatlib-0.2.13 → autostatlib-0.2.16}/src/AutoStatLib/StatPlots.py

@@ -1,3 +1,4 @@
+import seaborn as sns
 import random
 # from math import comb
 import numpy as np
@@ -87,6 +88,9 @@ class BaseStatPlot(Helpers):
                  colormap=None,
                  print_p_label=True,
                  print_stars=True,
+                 figure_scale_factor=1,
+                 figure_h=4,
+                 figure_w=0,  # 0 means auto
                  **kwargs):
         self.data_groups = [group if group else [0, 0, 0, 0]
                             for group in data_groups]
@@ -103,6 +107,9 @@ class BaseStatPlot(Helpers):
         self.print_p_label = print_p_label
         self.print_stars = print_stars
         self.print_x_labels = print_x_labels
+        self.figure_scale_factor = figure_scale_factor
+        self.figure_h = figure_h
+        self.figure_w = figure_w
 
         #  sd sem mean and median calculation if they are not provided
         self.mean = [
@@ -132,31 +139,16 @@ class BaseStatPlot(Helpers):
         self.y_max = max([max(data) for data in self.data_groups])
 
     def setup_figure(self, ):
-        fig, ax = plt.subplots(figsize=(0.5 + 0.9 * self.n_groups, 4))
-        return fig, ax
+        fig, ax = plt.subplots(
+            dpi=100,
+            figsize=((0.5 + 0.9 * self.n_groups)
+                     if not self.figure_w else self.figure_w, self.figure_h)
+        )
 
-    def add_scatter(self, ax,
-                    color='k',
-                    alpha=0.5,
-                    marker='o',
-                    linewidth=1,
-                    zorder=1):
-        # Generate x jitter pool.
-        spread_pool = []  # storing x positions of data points
-        for i, data in enumerate(self.data_groups):
-            spread = tuple(random.uniform(-.10, .10) for _ in data)
-            spread_pool.append(tuple(i + s for s in spread))
+        figure_size = plt.gcf().get_size_inches()
+        plt.gcf().set_size_inches(self.figure_scale_factor * figure_size)
 
-        for i, data in enumerate(self.transpose(self.data_groups)):
-            # Plot individual data points with x jitter.
-            ax.plot(self.transpose(spread_pool)[i], data,
-                    color=color,
-                    alpha=alpha,
-                    marker=marker,
-                    linewidth=linewidth,
-                    # Connect the data points if desired.
-                    linestyle='-' if self.dependent else '',
-                    zorder=zorder)
+        return fig, ax
 
     def add_barplot(self, ax, x,
                     fill=True,
@@ -169,7 +161,7 @@ class BaseStatPlot(Helpers):
                facecolor=self.colors_fill[x % len(self.colors_fill)],
                edgecolor=self.colors_edge[x % len(self.colors_edge)],
                fill=fill,
-               linewidth=linewidth,
+               linewidth=linewidth*self.figure_scale_factor,
                zorder=zorder)
 
     def add_violinplot(self, ax, x,
@@ -189,7 +181,7 @@ class BaseStatPlot(Helpers):
         for pc in vp['bodies']:
             pc.set_facecolor(self.colors_fill[x % len(self.colors_fill)])
             pc.set_edgecolor(self.colors_edge[x % len(self.colors_edge)])
-            pc.set_linewidth(linewidth)
+            pc.set_linewidth(linewidth*self.figure_scale_factor)
 
     def add_boxplot(self, ax,
                     # positions of boxes, defaults to range(1,n+1)
@@ -343,6 +335,83 @@ class BaseStatPlot(Helpers):
                 patch.set_facecolor(
                     self.colors_fill[x % len(self.colors_fill)])
 
+    def add_scatter(self, ax,
+                    color='k',
+                    alpha=0.5,
+                    marker='o',
+                    markersize=8,
+                    linewidth=1.2,
+                    zorder=1):
+        # Generate x jitter pool.
+        spread_pool = []  # storing x positions of data points
+        for i, data in enumerate(self.data_groups):
+            spread = tuple(random.uniform(-.10, .10) for _ in data)
+            spread_pool.append(tuple(i + s for s in spread))
+
+        for i, data in enumerate(self.transpose(self.data_groups)):
+            # Plot individual data points with x jitter.
+            ax.plot(self.transpose(spread_pool)[i], data,
+                    color=color,
+                    alpha=alpha,
+                    marker=marker,
+                    markersize=markersize*self.figure_scale_factor,
+                    linewidth=linewidth*self.figure_scale_factor,
+                    # Connect the data points if desired.
+                    linestyle='-' if self.dependent else '',
+                    zorder=zorder)
+
+    def add_swarm(self, ax,
+                  color='grey',
+                  alpha=1,
+                  marker='o',
+                  markersize=8,
+                  linewidth=1.4,
+                  zorder=1):
+        """
+        Add a swarmplot (scatter-like plot with non-overlapping points)
+        to the provided Axes. Automatically reduce point size if overcrowded.
+        """
+
+        # Prepare flattened data
+        values = [v for i, group in enumerate(self.data_groups) for v in group]
+        groups = [i for i, group in enumerate(self.data_groups) for _ in group]
+
+        # Estimate overcrowding for adaptive sizing
+        group_counts = [len(g) for g in self.data_groups]
+        max_points = max(group_counts) if group_counts else 1
+
+        # Determine horizontal space per category
+        num_groups = len(self.data_groups)
+        xlim = ax.get_xlim()
+        width_per_group = (xlim[1] - xlim[0]) / max(num_groups, 1)
+
+        # Empirical density threshold: if points are too dense, shrink
+        density = max_points / (width_per_group + 1e-6)
+
+        # Tunable constants to approximate best function of size adjustment
+        size_scale = max(0.1, min(1, 3.5 / (density ** 0.5)))
+
+        sns.swarmplot(
+            x=groups,
+            y=values,
+            ax=ax,
+            color=color,
+            alpha=alpha,
+            size=markersize * self.figure_scale_factor * size_scale,
+            marker=marker,
+            linewidth=linewidth * self.figure_scale_factor * size_scale,
+            zorder=zorder,
+        )
+
+        # Connect points if data paired
+        if self.dependent == True:
+            for i, data in enumerate(self.transpose(self.data_groups)):
+                ax.plot(range(len(data)), data,
+                        color=color,
+                        alpha=alpha * 0.7,
+                        linewidth=linewidth * self.figure_scale_factor,
+                        zorder=zorder - 1)
+
     def add_errorbar_sd(self, ax, x,
                         capsize=8,
                         ecolor='r',
@@ -352,9 +421,9 @@ class BaseStatPlot(Helpers):
         ax.errorbar(x, self.mean[x],
                     yerr=self.sd[x],
                     fmt='none',
-                    capsize=capsize,
+                    capsize=capsize*self.figure_scale_factor,
                     ecolor=ecolor,
-                    linewidth=linewidth,
+                    linewidth=linewidth*self.figure_scale_factor,
                     zorder=zorder)
 
     def add_errorbar_sem(self, ax, x,
@@ -366,9 +435,10 @@ class BaseStatPlot(Helpers):
         ax.errorbar(x, self.mean[x],
                     yerr=self.sem[x],
                     fmt='none',
-                    capsize=capsize,
+                    capsize=capsize*self.figure_scale_factor,
                     ecolor=ecolor,
-                    linewidth=linewidth,
+                    linewidth=linewidth*self.figure_scale_factor,
+                    elinewidth=linewidth*self.figure_scale_factor,
                     zorder=zorder)
 
     def add_mean_marker(self, ax, x,
@@ -382,8 +452,8 @@ class BaseStatPlot(Helpers):
                 marker=marker,
                 markerfacecolor=markerfacecolor,
                 markeredgecolor=markeredgecolor,
-                markersize=markersize,
-                markeredgewidth=markeredgewidth)
+                markersize=markersize*self.figure_scale_factor,
+                markeredgewidth=markeredgewidth*self.figure_scale_factor)
 
     def add_median_marker(self, ax, x,
                           marker='x',
@@ -396,8 +466,8 @@ class BaseStatPlot(Helpers):
                 marker=marker,
                 markerfacecolor=markerfacecolor,
                 markeredgecolor=markeredgecolor,
-                markersize=markersize,
-                markeredgewidth=markeredgewidth)
+                markersize=markersize*self.figure_scale_factor,
+                markeredgewidth=markeredgewidth*self.figure_scale_factor)
 
     def add_significance_bars(self, ax,
                               linewidth=2,
@@ -417,13 +487,13 @@ class BaseStatPlot(Helpers):
 
             match (self.print_p_label, self.print_stars):
                 case (True, True):
-                    vspace = capsize+0.06
+                    vspace = (capsize+0.06)*self.figure_scale_factor
                     label = '{}\n{}'.format(p, stars)
                 case (True, False):
-                    vspace = capsize+0.03
+                    vspace = (capsize+0.03)*self.figure_scale_factor
                     label = '{}'.format(p)
                 case (False, True):
-                    vspace = capsize+0.03
+                    vspace = (capsize+0.03)*self.figure_scale_factor
                     label = '{}'.format(stars)
 
             if self.print_p_label or self.print_stars:
@@ -431,10 +501,10 @@ class BaseStatPlot(Helpers):
                 y, h = ((1.05 + (order*vspace)) *
                         self.y_max), capsize * self.y_max
                 ax.plot([x1, x1, x2, x2], [y, y + h, y + h, y],
-                        lw=linewidth, c=col)
+                        lw=linewidth*self.figure_scale_factor, c=col)
 
                 ax.text((x1 + x2) * 0.5, y + h, label,
-                        ha='center', va='bottom', color=col, fontweight='bold', fontsize=8)
+                        ha='center', va='bottom', color=col, fontweight='bold', fontsize=8*self.figure_scale_factor)
 
         def draw_bar_from_posthoc_matrix(x1, x2, o):
             draw_bar(
@@ -506,7 +576,7 @@ class BaseStatPlot(Helpers):
             else:
                 ax.set_xticks(range(self.n_groups))
                 ax.set_xticklabels(['Group {}'.format(i + 1)
-                                   for i in range(self.n_groups)], fontweight='regular', fontsize=8)
+                                   for i in range(self.n_groups)], fontweight='regular', fontsize=8*self.figure_scale_factor)
         else:
             plt.subplots_adjust(bottom=0.08)
             ax.tick_params(axis='x', which='both',
@@ -515,31 +585,37 @@ class BaseStatPlot(Helpers):
         # Additional formatting
         for ytick in ax.get_yticklabels():
             ytick.set_fontweight('bold')
-        ax.tick_params(width=linewidth)
-        ax.xaxis.set_tick_params(labelsize=10)
-        ax.yaxis.set_tick_params(labelsize=12)
-        ax.spines['left'].set_linewidth(linewidth)
+        ax.tick_params(width=linewidth*self.figure_scale_factor)
+        ax.xaxis.set_tick_params(labelsize=10*self.figure_scale_factor)
+        ax.yaxis.set_tick_params(labelsize=12*self.figure_scale_factor)
+        ax.spines['left'].set_linewidth(linewidth*self.figure_scale_factor)
         ax.tick_params(axis='y', which='both',
-                       length=linewidth * 2, width=linewidth)
+                       length=linewidth * 2*self.figure_scale_factor, width=linewidth*self.figure_scale_factor)
         ax.tick_params(axis='x', which='both', length=0)
 
     def add_titles_and_labels(self, fig, ax):
         if self.plot_title:
-            ax.set_title(self.plot_title, fontsize=12, fontweight='bold')
+            ax.set_title(self.plot_title, fontsize=12 *
+                         self.figure_scale_factor, fontweight='bold')
         if self.x_label:
-            ax.set_xlabel(self.x_label, fontsize=10, fontweight='bold')
+            ax.set_xlabel(self.x_label, fontsize=10 *
+                          self.figure_scale_factor, fontweight='bold')
         if self.y_label:
-            ax.set_ylabel(self.y_label, fontsize=10, fontweight='bold')
+            ax.set_ylabel(self.y_label, fontsize=10 *
+                          self.figure_scale_factor, fontweight='bold')
         fig.text(0.95, 0.0,
-                 '{}, {}\nn={}'.format(self.testname, self.posthoc_name,
-                                       str(self.n)[1:-1] if not self.dependent else str(self.n[0])),
-                 ha='right', va='bottom', fontsize=8, fontweight='regular')
+                 '{}{}\nn={}'.format(self.testname, (', ' + self.posthoc_name) if self.posthoc_name else '',
+                                     str(self.n)[1:-1] if not self.dependent else str(self.n[0])),
+                 ha='right', va='bottom', fontsize=8*self.figure_scale_factor, fontweight='regular')
 
     def show(self):
         plt.show()
 
     def save(self, path):
-        plt.savefig(path)
+        plt.savefig(path,
+                    pad_inches=0.1*self.figure_scale_factor,
+                    transparent=True,
+                    )
 
     def close(self):
         plt.close()
@@ -566,7 +642,7 @@ class BarStatPlot(BaseStatPlot):
             self.add_mean_marker(ax, x)
             self.add_errorbar_sd(ax, x)
 
-        self.add_scatter(ax)
+        self.add_swarm(ax)
         self.add_significance_bars(ax, linewidth)
         self.add_titles_and_labels(fig, ax)
         self.axes_formatting(ax, linewidth)
@@ -599,7 +675,7 @@ class ViolinStatPlot(BaseStatPlot):
             self.add_mean_marker(ax, x)
             # self.add_errorbar_sd(ax, x)
 
-        self.add_scatter(ax)
+        self.add_swarm(ax)
         self.add_significance_bars(ax, linewidth)
         self.add_titles_and_labels(fig, ax)
         self.axes_formatting(ax, linewidth)
@@ -612,7 +688,7 @@ class BoxStatPlot(BaseStatPlot):
         linewidth = 2
 
         self.add_boxplot(ax)
-        self.add_scatter(ax)
+        self.add_swarm(ax)
         self.add_significance_bars(ax, linewidth)
         self.add_titles_and_labels(fig, ax)
         self.axes_formatting(ax, linewidth)
@@ -635,3 +711,28 @@ class ScatterStatPlot(BaseStatPlot):
         self.add_significance_bars(ax, linewidth)
         self.add_titles_and_labels(fig, ax)
         self.axes_formatting(ax, linewidth)
+
+        xmin, xmax = ax.get_xlim()
+        ax.set_xlim(xmin - 0.25, xmax + 0.25)
+
+
+class SwarmStatPlot(BaseStatPlot):
+
+    def plot(self):
+        fig, ax = self.setup_figure()
+        linewidth = 2
+
+        for x in range(len(self.data_groups)):
+
+            # Overlay errbars, and markers.
+            self.add_median_marker(ax, x)
+            self.add_mean_marker(ax, x)
+            self.add_errorbar_sd(ax, x)
+
+        self.add_swarm(ax)
+        self.add_significance_bars(ax, linewidth)
+        self.add_titles_and_labels(fig, ax)
+        self.axes_formatting(ax, linewidth)
+
+        xmin, xmax = ax.get_xlim()
+        ax.set_xlim(xmin - 0.25, xmax + 0.25)

{autostatlib-0.2.13 → autostatlib-0.2.16}/src/AutoStatLib/_version.py

@@ -1,2 +1,2 @@
 # AutoStatLib package version:
-__version__ = "0.2.13"
+__version__ = "0.2.16"

{autostatlib-0.2.13 → autostatlib-0.2.16/src/AutoStatLib.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: AutoStatLib
-Version: 0.2.13
+Version: 0.2.16
 Summary: AutoStatLib - a simple statistical analysis tool
 Author: Stemonitis, SciWare LLC
 Author-email: konung-yaropolk <yaropolk1995@gmail.com>
@@ -533,6 +533,7 @@ Requires-Dist: numpy
 Requires-Dist: scipy
 Requires-Dist: statsmodels
 Requires-Dist: matplotlib
+Requires-Dist: seaborn
 Requires-Dist: scikit-posthocs
 Requires-Dist: pandas
 Dynamic: license-file
@@ -675,22 +676,22 @@ If errors occured, *GetResult()* returns an empty dictionary
 -- add QQ plot
 -- n-sample tests: add onetail option
 
-✅ done -- detailed normality test results
-✅ done -- added posthoc: Kruskal-Wallis Dunn's multiple comparisons
-
-tests check:
-1-sample:
---Wilcoxon 2,1 tails - ✅ok
---t-tests 2,1 tails -✅ok
-
-2-sample:
---Wilcoxon 2,1 tails - ✅ok
---Mann-whitney 2,1 tails - ✅ok
---t-tests 2,1 tails -✅ok
-
-n-sample:
---Kruskal-Wallis 2 tail - ✅ok
---Dunn's multiple comparisons - ✅ok
---Friedman 2 tail - ✅ok
---one-way ANOWA 2 tail - ✅ok
---Tukey`s multiple comparisons - ✅ok
+✅ done -- detailed normality test results  
+✅ done -- added posthoc: Kruskal-Wallis Dunn's multiple comparisons  
+
+tests check:  
+1-sample:  
+--Wilcoxon 2,1 tails - ✅ok  
+--t-tests 2,1 tails -✅ok  
+
+2-sample:  
+--Wilcoxon 2,1 tails - ✅ok  
+--Mann-whitney 2,1 tails - ✅ok  
+--t-tests 2,1 tails -✅ok  
+
+n-sample:  
+--Kruskal-Wallis 2 tail - ✅ok  
+--Dunn's multiple comparisons - ✅ok   
+--Friedman 2 tail - ✅ok  
+--one-way ANOWA 2 tail - ✅ok  
+--Tukey`s multiple comparisons - ✅ok  

{autostatlib-0.2.13 → autostatlib-0.2.16}/src/AutoStatLib.egg-info/requires.txt

@@ -2,5 +2,6 @@ numpy
 scipy
 statsmodels
 matplotlib
+seaborn
 scikit-posthocs
 pandas