AutoStatLib 0.4.0__tar.gz → 0.4.2__tar.gz

{autostatlib-0.4.0/src/AutoStatLib.egg-info → autostatlib-0.4.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: AutoStatLib
-Version: 0.4.0
+Version: 0.4.2
 Summary: AutoStatLib - a simple statistical analysis tool
 Author: Stemonitis, SciWare LLC
 Author-email: konung-yaropolk <yaropolk1995@gmail.com>

{autostatlib-0.4.0 → autostatlib-0.4.2}/src/AutoStatLib/AutoStatLib.py

@@ -1,5 +1,6 @@
 from __future__ import annotations
 
+from itertools import cycle, islice
 from typing import Optional, Union
 
 import numpy as np
@@ -69,7 +70,7 @@ class StatisticalAnalysis(StatisticalTests, NormalityTests, TextFormatting, Help
         self.raise_errors: bool = raise_errors
         self.n_groups: int = len(self.groups_list)
         self.groups_name: list[str] = (
-            [groups_name[i % len(groups_name)] for i in range(self.n_groups)]
+            list(islice(cycle(groups_name), self.n_groups))
             if groups_name and groups_name != [""]
             else [f"Group {i + 1}" for i in range(self.n_groups)]
         )
@@ -216,7 +217,7 @@ class StatisticalAnalysis(StatisticalTests, NormalityTests, TextFormatting, Help
             ), "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"
+            ), "Each group must be at least n=4 for the valid statistics, so test was skipped"
             assert not (
                 test in self.test_ids_dependent
                 and not all(len(lst) == len(self.data[0]) for lst in self.data)

{autostatlib-0.4.0 → autostatlib-0.4.2}/src/AutoStatLib/StatPlots.py

@@ -20,12 +20,8 @@ class Helpers:
         colors: list[str | tuple],
         alpha: float = 0.35,
     ) -> list[tuple[float, float, float, float]]:
-        rgba_colors: list[tuple[float, float, float, float]] = []
-        for col in colors:
-            rgba = list(mcolors.to_rgba(col))
-            rgba[3] = alpha
-            rgba_colors.append((rgba[0], rgba[1], rgba[2], rgba[3]))
-        return rgba_colors
+        # mcolors.to_rgba returns a 4-tuple; replace only the alpha channel.
+        return [(*mcolors.to_rgba(c)[:3], alpha) for c in colors]
 
     def get_colors(
         self,
@@ -166,26 +162,34 @@ class BaseStatPlot(Helpers):
             print("AutoStatLib.StatPlots Error :", error)
             return
 
-        #  sd sem mean and median calculation if they are not provided
-        self.mean: list[float] = [
-            np.mean(self.data_groups[i]).item() for i in range(self.n_groups)
-        ]
-        self.median: list[float] = [
-            np.median(self.data_groups[i]).item() for i in range(self.n_groups)
-        ]
-        self.sd: list[float] = [
-            np.std(self.data_groups[i], ddof=1).item() for i in range(self.n_groups)
-        ]
-        self.sem: list[float] = [
-            np.std(self.data_groups[i], ddof=1).item()
-            / np.sqrt(len(self.data_groups[i]))
-            for i in range(self.n_groups)
-        ]
-
-        self.n: list[int] = [len(i) for i in self.data_groups]
+        # sd sem mean and median calculation if they are not provided.
+        # Convert each group to a float array once; reuse for all four stats.
+        # This avoids calling np.std twice per group (old code recomputed it
+        # from scratch for sem after already computing it for sd).
+        _arrs       = [np.asarray(g, dtype=float) for g in self.data_groups]
+        self.n      = [len(a) for a in _arrs]
+        self.mean   = [float(a.mean())      for a in _arrs]
+        self.median = [float(np.median(a))  for a in _arrs]
+        self.sd     = [float(a.std(ddof=1)) for a in _arrs]
+        self.sem    = [sd / np.sqrt(n) for sd, n in zip(self.sd, self.n)]
         self.p_printed: str = self.make_p_value_printed(self.p)
         self.stars_printed: str = self.make_stars_printed(self.make_stars(self.p))
 
+        # Pre-compute posthoc matrix string representations once here so that
+        # add_significance_bars() doesn't rebuild them on every call.
+        if self.posthoc_matrix:
+            self._posthoc_printed: list[list[str]] = [
+                [self.make_p_value_printed(e) for e in row]
+                for row in self.posthoc_matrix
+            ]
+            self._posthoc_stars: list[list[str]] = [
+                [self.make_stars_printed(self.make_stars(e)) for e in row]
+                for row in self.posthoc_matrix
+            ]
+        else:
+            self._posthoc_printed = []
+            self._posthoc_stars   = []
+
         self.groups_name: list[str] = Groups_Name if Groups_Name is not None else [""]
         self.subgrouping: list = subgrouping if subgrouping else [0]
         self.subgrouping_arrange: list[int] = self.expand_counts(self.subgrouping)
@@ -430,10 +434,12 @@ class BaseStatPlot(Helpers):
         linewidth: float = 1.2,
         zorder: int = 2,
     ) -> None:
-        spread_pool: list[tuple] = []
-        for i, data in enumerate(self.data_groups):
-            spread = tuple(random.uniform(-0.10, 0.10) for _ in data)
-            spread_pool.append(tuple(i + s for s in spread))
+        # Generate all jitter offsets with NumPy at once instead of Python loops.
+        rng = np.random.default_rng()
+        spread_pool: list[np.ndarray] = [
+            i + rng.uniform(-0.10, 0.10, size=len(g))
+            for i, g in enumerate(self.data_groups)
+        ]
 
         for i, data in enumerate(self.transpose(self.data_groups)):
             ax.plot(
@@ -490,6 +496,7 @@ class BaseStatPlot(Helpers):
             marker=marker,
             linewidth=linewidth * self.figure_scale_factor * size_scale,
             zorder=zorder,
+            warn_thresh = 1, # threshold for warning about too many points; set to 0 to always warn, or 1 to never warn
         )
 
         if self.dependent:
@@ -690,22 +697,10 @@ class BaseStatPlot(Helpers):
         col: str = "k",
     ) -> None:
 
-        posthoc_matrix_printed: list[list[str]] = (
-            [
-                [self.make_p_value_printed(element) for element in row]
-                for row in self.posthoc_matrix
-            ]
-            if self.posthoc_matrix
-            else []
-        )
-        posthoc_matrix_stars: list[list[str]] = (
-            [
-                [self.make_stars_printed(self.make_stars(element)) for element in row]
-                for row in self.posthoc_matrix
-            ]
-            if self.posthoc_matrix
-            else []
-        )
+        # Use the pre-computed representations cached in __init__ rather than
+        # rebuilding them on every call to add_significance_bars().
+        posthoc_matrix_printed: list[list[str]] = self._posthoc_printed
+        posthoc_matrix_stars:   list[list[str]] = self._posthoc_stars
 
         def draw_bar(
             p: str,

{autostatlib-0.4.0 → autostatlib-0.4.2}/src/AutoStatLib/_version.py

@@ -1,2 +1,2 @@
 # AutoStatLib package version:
-__version__ = "0.4.0"
+__version__ = "0.4.2"

{autostatlib-0.4.0 → autostatlib-0.4.2}/src/AutoStatLib/helpers.py

@@ -11,17 +11,18 @@ import pandas as pd
 class Helpers(StatAnalysisProtocol):
 
     def matrix_to_dataframe(self, matrix: list[list[float]]) -> pd.DataFrame:
-        data: list[float] = []
-        cols: list[int] = []
-        rows: list[int] = []
-
-        for i, row in enumerate(matrix):
-            for j, value in enumerate(row):
-                data.append(value)
-                cols.append(i)
-                rows.append(j)
-
-        return pd.DataFrame({"Row": rows, "Col": cols, "Value": data})
+        # Convert once to a 2-D float array, then use NumPy meshgrid to build
+        # the row/col index arrays without any Python-level loop.
+        arr = np.array(matrix, dtype=float)          # (n_subjects, n_conditions)
+        n_rows, n_cols = arr.shape
+        row_idx, col_idx = np.meshgrid(
+            np.arange(n_rows), np.arange(n_cols), indexing="ij"
+        )
+        return pd.DataFrame({
+            "Row":   row_idx.ravel(),
+            "Col":   col_idx.ravel(),
+            "Value": arr.ravel(),
+        })
 
     def list_to_matrix(self, values: list[float], n: int) -> list[list[float]]:
         i = 0
@@ -65,6 +66,32 @@ class Helpers(StatAnalysisProtocol):
             self.make_stars_printed(self.stars_int) if self.successfull else ""
         )
 
+        # --- Compute per-group descriptive stats in a single pass ----------
+        # Convert each group once; reuse the array for mean, median, std, sem.
+        # This also avoids calling np.std twice (once for SD, once for SE).
+        groups_arr    = [np.asarray(g, dtype=float) for g in self.data]
+        groups_n      = [len(a)                         for a in groups_arr]
+        groups_mean   = [float(a.mean())                for a in groups_arr]
+        groups_median = [float(np.median(a))            for a in groups_arr]
+        groups_sd     = [float(a.std(ddof=1))           for a in groups_arr]
+        groups_se     = [sd / np.sqrt(n) for sd, n in zip(groups_sd, groups_n)]
+
+        # --- Posthoc matrix representations — one pass over the matrix -----
+        # Previously built as three separate nested list comprehensions;
+        # now all three are filled in a single traversal.
+        if self.posthoc_matrix:
+            pm_bool:    list[list] = []
+            pm_printed: list[list] = []
+            pm_stars:   list[list] = []
+            for row in self.posthoc_matrix:
+                pm_bool.append([bool(e) for e in row])
+                pm_printed.append([self.make_p_value_printed(e) for e in row])
+                pm_stars.append(
+                    [self.make_stars_printed(self.make_stars(e)) for e in row]
+                )
+        else:
+            pm_bool = pm_printed = pm_stars = []
+
         return {
             "p_value": (
                 self.make_p_value_printed(self.p_value.item())
@@ -90,52 +117,22 @@ class Helpers(StatAnalysisProtocol):
             "Stars": self.stars_int,
             "Warnings": self.warnings,
             "Successfull_Test": (self.successfull and not self.error),
-            "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))
-            ],
-            "Groups_SD": [
-                np.std(self.data[i], ddof=1).item() for i in range(len(self.data))
-            ],
-            "Groups_SE": [
-                np.std(self.data[i], ddof=1).item() / np.sqrt(len(self.data[i]))
-                for i in range(len(self.data))
-            ],
+            "Groups_Name":   self.groups_name,
+            "Groups_N":      groups_n,
+            "Groups_Median": groups_median,
+            "Groups_Mean":   groups_mean,
+            "Groups_SD":     groups_sd,
+            "Groups_SE":     groups_se,
             "subgrouping": self.subgrouping,
             # actually returns list of lists of numpy dtypes of float64, next make it return regular floats:
             "Samples": self.data,
             "Posthoc_Tests_Name": (
                 self.posthoc_name if self.posthoc_name is not None else ""
             ),
-            "Posthoc_Matrix": self.posthoc_matrix if self.posthoc_matrix else [],
-            "Posthoc_Matrix_bool": (
-                [[bool(element) for element in row] for row in self.posthoc_matrix]
-                if self.posthoc_matrix
-                else []
-            ),
-            "Posthoc_Matrix_printed": (
-                [
-                    [self.make_p_value_printed(element) for element in row]
-                    for row in self.posthoc_matrix
-                ]
-                if self.posthoc_matrix
-                else []
-            ),
-            "Posthoc_Matrix_stars": (
-                [
-                    [
-                        self.make_stars_printed(self.make_stars(element))
-                        for element in row
-                    ]
-                    for row in self.posthoc_matrix
-                ]
-                if self.posthoc_matrix
-                else []
-            ),
+            "Posthoc_Matrix":         self.posthoc_matrix if self.posthoc_matrix else [],
+            "Posthoc_Matrix_bool":    pm_bool,
+            "Posthoc_Matrix_printed": pm_printed,
+            "Posthoc_Matrix_stars":   pm_stars,
         }
 
     def log(self, *args: object, **kwargs: object) -> None:
@@ -145,4 +142,4 @@ class Helpers(StatAnalysisProtocol):
     def AddWarning(self, warning_id: str) -> None:
         message: str = self.warning_ids_all[warning_id]
         self.log(message)
-        self.warnings.append(message)
+        self.warnings.append(message)

{autostatlib-0.4.0 → autostatlib-0.4.2}/src/AutoStatLib/statistical_tests.py

@@ -241,8 +241,8 @@ class StatisticalTests(StatAnalysisProtocol):
         if self.popmean is None:
             self.popmean = 0
             self.AddWarning("no_pop_mean_set")
-        data: list[float] = [x - self.popmean for x in self.data[0]]
-        stat, p_value = wilcoxon(data)
+        arr = np.asarray(self.data[0], dtype=float) - self.popmean
+        stat, p_value = wilcoxon(arr)
         if self.tails == 1:
             p_value /= 2
         return stat, p_value

{autostatlib-0.4.0 → autostatlib-0.4.2}/src/AutoStatLib/text_formatting.py

@@ -8,27 +8,29 @@ from typing import Optional
 class TextFormatting(StatAnalysisProtocol):
     """Text formatting mixin."""
 
-    def autospace(
-        self, elements_list: list[str], space: int, delimiter: str = " "
-    ) -> str:
-        output = ""
-        for i, element in enumerate(elements_list):
-            if i == len(elements_list):
-                output += element
-            else:
-                output += element + (space - len(element)) * delimiter
-        return output
+    def _fmt_row(self, elements: list[str], width: int, fill: str = " ") -> str:
+        """
+        Format a list of strings into a fixed-width columnar row.
+
+        Each element is left-justified to ``width`` characters using ``fill``
+        as the pad character.  The last element is appended without trailing
+        padding (matches terminal/log output intent).
+
+        Replaces the hand-rolled ``autospace()`` loop with Python's built-in
+        ``str.ljust`` and ``str.join``.
+        """
+        if not elements:
+            return ""
+        # All but the last element are padded to `width`; last is bare.
+        return "".join(e.ljust(width, fill) for e in elements[:-1]) + elements[-1]
 
     def print_groups(self, space: int = 24, max_length: int = 15) -> None:
         self.log("")
         data: list[list[float]] = self.data
-        num_groups: int = len(data)
         group_longest: int = max(len(row) for row in data)
 
-        header: list[str] = self.groups_name
-        line: list[str] = ["" * 7]
-        self.log(self.autospace(header, space))
-        self.log(self.autospace(line, space))
+        self.log(self._fmt_row(self.groups_name, space))
+        self.log(self._fmt_row(["" * 7], space))
 
         for i in range(group_longest):
             row_values: list[str] = []
@@ -51,7 +53,7 @@ class TextFormatting(StatAnalysisProtocol):
                         row_values.append("")
             if all_values_empty:
                 break
-            self.log(self.autospace(row_values, space))
+            self.log(self._fmt_row(row_values, space))
 
     def print_results(self) -> None:
         self.log("\n\nResults: \n")

{autostatlib-0.4.0 → autostatlib-0.4.2/src/AutoStatLib.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: AutoStatLib
-Version: 0.4.0
+Version: 0.4.2
 Summary: AutoStatLib - a simple statistical analysis tool
 Author: Stemonitis, SciWare LLC
 Author-email: konung-yaropolk <yaropolk1995@gmail.com>