pandas-plots 0.15.10__tar.gz → 0.15.13__tar.gz

{pandas_plots-0.15.10 → pandas_plots-0.15.13}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.3
 Name: pandas-plots
-Version: 0.15.10
+Version: 0.15.13
 Summary: A collection of helper for table handling and visualization
 Keywords: tables,pivot,plotly,venn,plot,vizualization
 Author: smeisegeier

{pandas_plots-0.15.10 → pandas_plots-0.15.13}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "pandas-plots"
-version = "0.15.10"
+version = "0.15.13"
 description = "A collection of helper for table handling and visualization"
 long_description = "file: README.md"
 long_description_content_type = "text/markdown"

{pandas_plots-0.15.10 → pandas_plots-0.15.13}/src/pandas_plots/hlp.py

@@ -21,49 +21,44 @@ from PIL import Image
 
 URL_REGEX = r"^(?:http|ftp)s?://"  # https://stackoverflow.com/a/1617386
 
-def mean_confidence_interval(data, confidence=0.95, use_median=False, n_bootstraps=1000):
+
+def mean_confidence_interval(df, confidence=0.95, use_median=False):
     """
-    Calculate the mean or median and confidence interval.
-    For median, uses bootstrapping for a more robust confidence interval.
+    Calculate the mean or median and confidence interval of the input dataframe.
+    Source: https://stackoverflow.com/questions/15033511/compute-a-confidence-interval-from-sample-data
 
     Parameters:
-    data (array-like): The input data.
+    df (array-like): The input dataframe.
     confidence (float, optional): The confidence level for the interval. Defaults to 0.95.
-    use_median (bool, optional): If True, calculates median and its confidence interval. Defaults to False.
-    n_bootstraps (int, optional): Number of bootstrap samples for median CI. Only used if use_median is True.
+    use_median (bool, optional): If True, calculates median and confidence interval instead of mean. Defaults to False.
 
     Returns:
-    tuple: A tuple containing the central value (mean or median), margin of error, lower bound, and upper bound.
+    tuple: A tuple containing the central value (mean or median), interval, lower bound, and upper bound.
     """
-    data = to_series(data)
-    if data is None or len(data) == 0:
-        return np.nan, np.nan, np.nan, np.nan
-    a = 1.0 * np.array(data)
+    df = to_series(df)
+    if df is None:
+        return None
+    a = 1.0 * np.array(df)
     n = len(a)
 
     if use_median:
-        if n < 2: # Cannot bootstrap with n < 2
-            return np.median(a), np.nan, np.nan, np.nan
-
-        bootstrapped_medians = []
-        for _ in range(n_bootstraps):
-            sample = np.random.choice(a, size=n, replace=True)
-            bootstrapped_medians.append(np.median(sample))
-
         median = np.median(a)
-        alpha = (1 - confidence) / 2
-        lower_bound = np.percentile(bootstrapped_medians, alpha * 100)
-        upper_bound = np.percentile(bootstrapped_medians, (1 - alpha) * 100)
-        margin = (upper_bound - lower_bound) / 2 # Simple approximation for margin based on interval width
-        return median, margin, lower_bound, upper_bound
+        se = 1.253 * scipy.stats.sem(a)  # Approximate standard error for median
+        margin = se * scipy.stats.t.ppf((1 + confidence) / 2.0, n - 1)
+        return median, margin, median - margin, median + margin
     else:
-        mean = np.mean(a)
-        if n <= 1:
-            return mean, np.nan, np.nan, np.nan
-        se = scipy.stats.sem(a)
+        mean, se = np.mean(a), scipy.stats.sem(a)
         margin = se * scipy.stats.t.ppf((1 + confidence) / 2.0, n - 1)
         return mean, margin, mean - margin, mean + margin
 
+    # # * Alternative
+    # # from statistics import NormalDist
+    # def confidence_interval(data, confidence=0.95):
+    #     dist = NormalDist.from_samples(data)
+    #     z = NormalDist().inv_cdf((1 + confidence) / 2.)
+    #     h = dist.stdev * z / ((len(data) - 1) ** .5)
+    #     return dist.mean - h, dist.mean + h
+
 
 def to_series(df) -> pd.Series | None:
     """

{pandas_plots-0.15.10 → pandas_plots-0.15.13}/src/pandas_plots/tbl.py

@@ -112,7 +112,7 @@ def describe_df(
         df=df,
         caption="dataframe",
         use_plot=True,
-        renderer="png",
+        renderer=None,
         template="plotly",
         fig_cols=3,
         fig_offset=None,
@@ -264,7 +264,7 @@ def describe_df(
         )
 
         fig.show(
-            renderer,
+            renderer=renderer or os.getenv("RENDERER"),
             width=fig_width * fig_cols,  # <-- Set width here
             height=fig_rowheight * fig_rows,  # <-- Set height here
         )
@@ -273,7 +273,6 @@ def describe_df(
         import missingno as msno
         msno.matrix(df_, figsize=missing_figsize)
 
-
 def pivot_df(
     df: pd.DataFrame,
     dropna: bool = False,