AutoStatLib 0.2.16__tar.gz → 0.2.20__tar.gz

{autostatlib-0.2.16/src/AutoStatLib.egg-info → autostatlib-0.2.20}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: AutoStatLib
-Version: 0.2.16
+Version: 0.2.20
 Summary: AutoStatLib - a simple statistical analysis tool
 Author: Stemonitis, SciWare LLC
 Author-email: konung-yaropolk <yaropolk1995@gmail.com>
@@ -659,39 +659,39 @@ If errors occured, *GetResult()* returns an empty dictionary
 
 ---
 
-## Pre-Alpha dev status.
+## Alpha dev status.
 
 ### TODO:
 
--- Anova: posthocs
--- Anova: add 2-way anova and 3-way anova
--- onevay Anova: add repeated measures (for normal dependent values) with and without Gaisser-Greenhouse correction
--- onevay Anova: add Brown-Forsithe and Welch (for normal independent values with unequal SDs between groups)
--- paired T-test: add ratio-paired t-test (ratios of paired values are consistent)
--- add Welch test (for norm data unequal variances)
--- add Kolmogorov-smirnov test (unpaired nonparametric 2 sample, compare cumulative distributions)
--- add independent t-test with Welch correction (do not assume equal SDs in groups)
--- add correlation test, correlation diagram
--- add linear regression, regression diagram
--- add QQ plot
--- n-sample tests: add onetail option
+-- Anova: posthocs  
+-- Anova: add 2-way anova and 3-way anova  
+-- onevay Anova: add repeated measures (for normal dependent values) with and without Gaisser-Greenhouse correction  
+-- onevay Anova: add Brown-Forsithe and Welch (for normal independent values with unequal SDs between groups)  
+-- paired T-test: add ratio-paired t-test (ratios of paired values are consistent)  
+-- add Welch test (for norm data unequal variances)  
+-- add Kolmogorov-smirnov test (unpaired nonparametric 2 sample, compare cumulative distributions)  
+-- add independent t-test with Welch correction (do not assume equal SDs in groups)  
+-- add correlation test, correlation diagram  
+-- add linear regression, regression diagram  
+-- 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  
+✅ok --Wilcoxon 2,1 tails  
+✅ok --t-tests 2,1 tails  
 
 2-sample:  
---Wilcoxon 2,1 tails - ✅ok  
---Mann-whitney 2,1 tails - ✅ok  
---t-tests 2,1 tails -✅ok  
+✅ok --Wilcoxon 2,1 tails  
+✅ok --Mann-whitney 2,1 tails  
+✅ok --t-tests 2,1 tails  
 
 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  
+✅ok --Kruskal-Wallis 2 tail  
+✅ok --Dunn's multiple comparisons  
+✅ok --Friedman 2 tail  
+✅ok --one-way ANOVA 2-tailed  
+✅ok --Tukey`s multiple comparisons  

{autostatlib-0.2.16 → autostatlib-0.2.20}/README.md

@@ -119,39 +119,39 @@ If errors occured, *GetResult()* returns an empty dictionary
 
 ---
 
-## Pre-Alpha dev status.
+## Alpha dev status.
 
 ### TODO:
 
--- Anova: posthocs
--- Anova: add 2-way anova and 3-way anova
--- onevay Anova: add repeated measures (for normal dependent values) with and without Gaisser-Greenhouse correction
--- onevay Anova: add Brown-Forsithe and Welch (for normal independent values with unequal SDs between groups)
--- paired T-test: add ratio-paired t-test (ratios of paired values are consistent)
--- add Welch test (for norm data unequal variances)
--- add Kolmogorov-smirnov test (unpaired nonparametric 2 sample, compare cumulative distributions)
--- add independent t-test with Welch correction (do not assume equal SDs in groups)
--- add correlation test, correlation diagram
--- add linear regression, regression diagram
--- add QQ plot
--- n-sample tests: add onetail option
+-- Anova: posthocs  
+-- Anova: add 2-way anova and 3-way anova  
+-- onevay Anova: add repeated measures (for normal dependent values) with and without Gaisser-Greenhouse correction  
+-- onevay Anova: add Brown-Forsithe and Welch (for normal independent values with unequal SDs between groups)  
+-- paired T-test: add ratio-paired t-test (ratios of paired values are consistent)  
+-- add Welch test (for norm data unequal variances)  
+-- add Kolmogorov-smirnov test (unpaired nonparametric 2 sample, compare cumulative distributions)  
+-- add independent t-test with Welch correction (do not assume equal SDs in groups)  
+-- add correlation test, correlation diagram  
+-- add linear regression, regression diagram  
+-- 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  
+✅ok --Wilcoxon 2,1 tails  
+✅ok --t-tests 2,1 tails  
 
 2-sample:  
---Wilcoxon 2,1 tails - ✅ok  
---Mann-whitney 2,1 tails - ✅ok  
---t-tests 2,1 tails -✅ok  
+✅ok --Wilcoxon 2,1 tails  
+✅ok --Mann-whitney 2,1 tails  
+✅ok --t-tests 2,1 tails  
 
 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  
+✅ok --Kruskal-Wallis 2 tail  
+✅ok --Dunn's multiple comparisons  
+✅ok --Friedman 2 tail  
+✅ok --one-way ANOVA 2-tailed  
+✅ok --Tukey`s multiple comparisons  

{autostatlib-0.2.16 → autostatlib-0.2.20}/src/AutoStatLib/AutoStatLib.py

@@ -18,7 +18,8 @@ class StatisticalAnalysis(StatisticalTests, NormalityTests, TextFormatting, Help
                  tails=2,
                  popmean=None,
                  posthoc=False,
-                 verbose=True):
+                 verbose=True,
+                 groups_name=[]):
         self.results = None
         self.error = False
         self.groups_list = groups_list
@@ -28,9 +29,28 @@ class StatisticalAnalysis(StatisticalTests, NormalityTests, TextFormatting, Help
         self.posthoc = posthoc
         self.verbose = verbose
         self.n_groups = len(self.groups_list)
+        self.groups_name = [groups_name[i % len(groups_name)]
+                             for i in range(self.n_groups)] if groups_name else [f'Group {i+1}'
+                                                                                       for i in range(self.n_groups)]
+
         self.warning_flag_non_numeric_data = False
         self.summary = 'AutoStatLib v{}'.format(__version__)
 
+        # empties
+        self.results = None
+        self.error = False
+        self.warnings = []
+        self.normals = []
+        self.test_name = ''
+        self.test_id = None
+        self.test_stat = None
+        self.p_value = None
+        self.posthoc_matrix_df = None
+        self.posthoc_matrix = []
+        self.posthoc_name = ''
+        self.data = []
+        self.parametric = None
+
         # test IDs classification:
         self.test_ids_all = [  # in aplhabetical order
             'anova_1w_ordinary',
@@ -80,6 +100,7 @@ class StatisticalAnalysis(StatisticalTests, NormalityTests, TextFormatting, Help
             'no_pop_mean_set':                 '\nWarning: No Population Mean was set up for single-sample test, used default 0 value.\n         The results might be skewed. \n         Please, set the Population Mean and run the test again.\n',
         }
 
+
     def run_test(self, test='auto'):
 
         # reset values from previous tests
@@ -108,7 +129,7 @@ class StatisticalAnalysis(StatisticalTests, NormalityTests, TextFormatting, Help
         # delete the empty cols from input
         self.data = [col for col in self.data if any(
             x is not None for x in col)]
-
+        
         # User input assertion block
         try:
             assert self.data, 'There is no input data'
@@ -116,7 +137,7 @@ class StatisticalAnalysis(StatisticalTests, NormalityTests, TextFormatting, Help
             assert test in self.test_ids_all or test == 'auto', 'Wrong test id choosen, ensure you called correct function'
             assert all(len(
                 group) >= 4 for group in self.data), 'Each group must contain at least four values'
-            assert not (self.paired == True
+            assert not (self.paired is True
                         and not all(len(lst) == len(self.data[0]) for lst in self.data)), 'Paired groups must have the same length'
             assert not (test in self.test_ids_dependent
                         and not all(len(lst) == len(self.data[0]) for lst in self.data)), 'Groups must have the same length for dependent groups test'
@@ -153,7 +174,7 @@ class StatisticalAnalysis(StatisticalTests, NormalityTests, TextFormatting, Help
                 '+' if x is True else '-' if x is False else ' ' if x is None else 'e' for x in poll)
             self.normals.append(isnormal)
             self.log(
-                f'        Group {i+1}:    {poll_print[0]}   {poll_print[1]}   {poll_print[2]}   {poll_print[3]}   so disrtibution seems {"normal" if isnormal else "not normal"}')
+                f'        {self.groups_name[i].ljust(7, ' ')[:7]}:    {poll_print[0]}   {poll_print[1]}   {poll_print[2]}   {poll_print[3]}   so disrtibution seems {"normal" if isnormal else "not normal"}')
         self.parametric = all(self.normals)
 
         # print test choosen
@@ -164,9 +185,9 @@ class StatisticalAnalysis(StatisticalTests, NormalityTests, TextFormatting, Help
         self.log('Test chosen by user:          ', test)
 
         # Wrong test Warnings
-        if not test == 'auto' and not self.parametric and test in self.test_ids_parametric:
+        if test != 'auto' and not self.parametric and test in self.test_ids_parametric:
             self.AddWarning('param_test_with_non-normal_data')
-        if not test == 'auto' and self.parametric and not test in self.test_ids_parametric:
+        if test != 'auto' and self.parametric and test not in self.test_ids_parametric:
             self.AddWarning('non-param_test_with_normal_data')
 
         # run the test
@@ -184,7 +205,7 @@ class StatisticalAnalysis(StatisticalTests, NormalityTests, TextFormatting, Help
         self.log('-'*67 + '\n')
 
         # print the results to console:
-        if self.verbose == True:
+        if self.verbose is True:
             print(self.summary)
 
     # public methods:

{autostatlib-0.2.16 → autostatlib-0.2.20}/src/AutoStatLib/StatPlots.py

@@ -82,7 +82,7 @@ class BaseStatPlot(Helpers):
                  x_label='',
                  y_label='',
                  print_x_labels=True,
-                 x_manual_tick_labels=None,
+                 Groups_Name=None,
                  Posthoc_Matrix=[],
                  Posthoc_Tests_Name='',
                  colormap=None,
@@ -125,7 +125,7 @@ class BaseStatPlot(Helpers):
         self.p_printed = self.make_p_value_printed(self.p)
         self.stars_printed = self.make_stars_printed(self.make_stars(self.p))
 
-        self.x_manual_tick_labels = x_manual_tick_labels if x_manual_tick_labels is not None else [
+        self.groups_name = Groups_Name if Groups_Name is not None else [
             '']
 
         if colormap is not None and colormap != ['']:
@@ -341,7 +341,7 @@ class BaseStatPlot(Helpers):
                     marker='o',
                     markersize=8,
                     linewidth=1.2,
-                    zorder=1):
+                    zorder=2):
         # Generate x jitter pool.
         spread_pool = []  # storing x positions of data points
         for i, data in enumerate(self.data_groups):
@@ -358,15 +358,15 @@ class BaseStatPlot(Helpers):
                     linewidth=linewidth*self.figure_scale_factor,
                     # Connect the data points if desired.
                     linestyle='-' if self.dependent else '',
-                    zorder=zorder)
+                    zorder=zorder-1)
 
     def add_swarm(self, ax,
-                  color='grey',
+                  color='dimgrey',
                   alpha=1,
                   marker='o',
                   markersize=8,
                   linewidth=1.4,
-                  zorder=1):
+                  zorder=2):
         """
         Add a swarmplot (scatter-like plot with non-overlapping points)
         to the provided Axes. Automatically reduce point size if overcrowded.
@@ -408,7 +408,7 @@ class BaseStatPlot(Helpers):
             for i, data in enumerate(self.transpose(self.data_groups)):
                 ax.plot(range(len(data)), data,
                         color=color,
-                        alpha=alpha * 0.7,
+                        alpha=alpha * 0.25,
                         linewidth=linewidth * self.figure_scale_factor,
                         zorder=zorder - 1)
 
@@ -569,10 +569,10 @@ class BaseStatPlot(Helpers):
         # Set x ticks and labels
         if self.print_x_labels:
             plt.subplots_adjust(bottom=0.11)
-            if self.x_manual_tick_labels != ['']:
+            if self.groups_name != ['']:
                 ax.set_xticks(range(self.n_groups))
-                ax.set_xticklabels([self.x_manual_tick_labels[i % len(self.x_manual_tick_labels)]
-                                    for i in range(self.n_groups)])
+                ax.set_xticklabels([self.groups_name[i % len(self.groups_name)]
+                                    for i in range(self.n_groups)], fontweight='regular', fontsize=8*self.figure_scale_factor)
             else:
                 ax.set_xticks(range(self.n_groups))
                 ax.set_xticklabels(['Group {}'.format(i + 1)
@@ -680,6 +680,9 @@ class ViolinStatPlot(BaseStatPlot):
         self.add_titles_and_labels(fig, ax)
         self.axes_formatting(ax, linewidth)
 
+        xmin, xmax = ax.get_xlim()
+        ax.set_xlim(xmin - 0.3, xmax + 0.3)
+
 
 class BoxStatPlot(BaseStatPlot):
 
@@ -713,7 +716,7 @@ class ScatterStatPlot(BaseStatPlot):
         self.axes_formatting(ax, linewidth)
 
         xmin, xmax = ax.get_xlim()
-        ax.set_xlim(xmin - 0.25, xmax + 0.25)
+        ax.set_xlim(xmin - 0.3, xmax + 0.3)
 
 
 class SwarmStatPlot(BaseStatPlot):
@@ -735,4 +738,4 @@ class SwarmStatPlot(BaseStatPlot):
         self.axes_formatting(ax, linewidth)
 
         xmin, xmax = ax.get_xlim()
-        ax.set_xlim(xmin - 0.25, xmax + 0.25)
+        ax.set_xlim(xmin - 0.3, xmax + 0.3)

{autostatlib-0.2.16 → autostatlib-0.2.20}/src/AutoStatLib/_version.py

@@ -1,2 +1,2 @@
 # AutoStatLib package version:
-__version__ = "0.2.16"
+__version__ = "0.2.20"

{autostatlib-0.2.16 → autostatlib-0.2.20}/src/AutoStatLib/helpers.py

@@ -21,6 +21,16 @@ class Helpers():
             {'Row': rows, 'Col': cols, 'Value': data})
         return df
 
+    def list_to_matrix(self, list, n):
+        i = 0
+        matrix = [[1.0 for _ in range(n)] for _ in range(n)]
+        for ax0 in range(n):
+            for ax1 in range(ax0+1, n):
+                matrix[ax0][ax1] = list[i]
+                matrix[ax1][ax0] = list[i]
+                i += 1
+        return matrix
+
     def floatify_recursive(self, data):
         if isinstance(data, list):
             # Recursively process sublists and filter out None values
@@ -55,6 +65,7 @@ class Helpers():
             'Stars':  self.stars_int,
             # 'Stat_Value': self.test_stat.item(),
             'Warnings': self.warnings,
+            'Groups_Name': self.groups_name,
             'Groups_N': [len(self.data[i]) for i in range(len(self.data))],
             'Groups_Median': [np.median(self.data[i]).item() for i in range(len(self.data))],
             'Groups_Mean': [np.mean(self.data[i]).item() for i in range(len(self.data))],

{autostatlib-0.2.16 → autostatlib-0.2.20}/src/AutoStatLib/statistical_tests.py

@@ -99,12 +99,12 @@ class StatisticalTests():
                 [[f"Group_{i+1}"] * len(group) for i, group in enumerate(self.data)])
             # Tukey's multiple comparisons
             tukey_result = pairwise_tukeyhsd(data_flat, group_labels)
-            p = tukey_result.pvalues.tolist()
-            self.posthoc_matrix = [
-                [p[2], p[0], p[1]],
-                [p[0], p[2], p[2]],
-                [p[1], p[2], p[2]],
-            ]
+
+            list = tukey_result.pvalues.tolist()
+            n = self.n_groups
+            # prepare posthoc matrix
+            self.posthoc_matrix = self.list_to_matrix(list, n)
+
         return stat, p_value
 
     def anova_1w_rm(self):

{autostatlib-0.2.16 → autostatlib-0.2.20}/src/AutoStatLib/text_formatting.py

@@ -22,7 +22,7 @@ class TextFormatting():
         group_longest = max(len(row) for row in data)
 
         # Print the header
-        header = [f'Group {i+1}' for i in range(num_groups)]
+        header = self.groups_name
         line = [''*7]
         self.log(self.autospace(header, space))
         self.log(self.autospace(line, space))

{autostatlib-0.2.16 → autostatlib-0.2.20/src/AutoStatLib.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: AutoStatLib
-Version: 0.2.16
+Version: 0.2.20
 Summary: AutoStatLib - a simple statistical analysis tool
 Author: Stemonitis, SciWare LLC
 Author-email: konung-yaropolk <yaropolk1995@gmail.com>
@@ -659,39 +659,39 @@ If errors occured, *GetResult()* returns an empty dictionary
 
 ---
 
-## Pre-Alpha dev status.
+## Alpha dev status.
 
 ### TODO:
 
--- Anova: posthocs
--- Anova: add 2-way anova and 3-way anova
--- onevay Anova: add repeated measures (for normal dependent values) with and without Gaisser-Greenhouse correction
--- onevay Anova: add Brown-Forsithe and Welch (for normal independent values with unequal SDs between groups)
--- paired T-test: add ratio-paired t-test (ratios of paired values are consistent)
--- add Welch test (for norm data unequal variances)
--- add Kolmogorov-smirnov test (unpaired nonparametric 2 sample, compare cumulative distributions)
--- add independent t-test with Welch correction (do not assume equal SDs in groups)
--- add correlation test, correlation diagram
--- add linear regression, regression diagram
--- add QQ plot
--- n-sample tests: add onetail option
+-- Anova: posthocs  
+-- Anova: add 2-way anova and 3-way anova  
+-- onevay Anova: add repeated measures (for normal dependent values) with and without Gaisser-Greenhouse correction  
+-- onevay Anova: add Brown-Forsithe and Welch (for normal independent values with unequal SDs between groups)  
+-- paired T-test: add ratio-paired t-test (ratios of paired values are consistent)  
+-- add Welch test (for norm data unequal variances)  
+-- add Kolmogorov-smirnov test (unpaired nonparametric 2 sample, compare cumulative distributions)  
+-- add independent t-test with Welch correction (do not assume equal SDs in groups)  
+-- add correlation test, correlation diagram  
+-- add linear regression, regression diagram  
+-- 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  
+✅ok --Wilcoxon 2,1 tails  
+✅ok --t-tests 2,1 tails  
 
 2-sample:  
---Wilcoxon 2,1 tails - ✅ok  
---Mann-whitney 2,1 tails - ✅ok  
---t-tests 2,1 tails -✅ok  
+✅ok --Wilcoxon 2,1 tails  
+✅ok --Mann-whitney 2,1 tails  
+✅ok --t-tests 2,1 tails  
 
 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  
+✅ok --Kruskal-Wallis 2 tail  
+✅ok --Dunn's multiple comparisons  
+✅ok --Friedman 2 tail  
+✅ok --one-way ANOVA 2-tailed  
+✅ok --Tukey`s multiple comparisons