pandas-plots 0.12.20__py3-none-any.whl → 0.12.22__py3-none-any.whl

pandas_plots/hlp.py

@@ -22,29 +22,34 @@ from PIL import Image
 URL_REGEX = r"^(?:http|ftp)s?://"  # https://stackoverflow.com/a/1617386
 
 
-def mean_confidence_interval(df, confidence=0.95):
+def mean_confidence_interval(df, confidence=0.95, use_median=False):
     """
-    Calculate the mean and confidence interval of the input dataframe.
-    source: https://stackoverflow.com/questions/15033511/compute-a-confidence-interval-from-sample-data
+    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:
     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 confidence interval instead of mean. Defaults to False.
 
     Returns:
-    tuple: A tuple containing the mean, interval, lower bound, and upper bound.
+    tuple: A tuple containing the central value (mean or median), interval, lower bound, and upper bound.
     """
     df = to_series(df)
     if df is None:
         return None
     a = 1.0 * np.array(df)
     n = len(a)
-    mean, se = np.mean(a), scipy.stats.sem(a)
-    # * calculate the margin of error for the confidence interval using the t-distribution with the specified confidence level.
-    margin = se * scipy.stats.t.ppf((1 + confidence) / 2.0, n - 1)
-    lower = mean - margin
-    upper = mean + margin
-    return mean, margin, lower, upper
+
+    if use_median:
+        median = np.median(a)
+        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, 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
@@ -190,7 +195,8 @@ def wrap_text(
         line = line + word_s + " "
         # * reset if counter exceeds limit, or if word ends with newline
         if i >= max_items_in_line or str(word).endswith("\n"):
-            out = out + line + "\n"
+            # out = out + line + "\n"
+            out = out + line.rstrip() + "  \n"
             line = ""
             i = 0
         # else:
@@ -542,7 +548,9 @@ def add_measures_to_pyg_config(json_path: str, nodes: list[tuple[str, str]] = [(
 
     Example
     -------
-    `add_measures_to_pyg_config('config.json', [('cnt_tum', 'count(distinct z_tum_id)')], strict=True)`
+    default: `add_measures_to_pyg_config('config.json', [('cnt_tum', 'count(distinct z_tum_id)')], strict=True)`
+    
+    usage: start pygwalker with empty config file but defined config path. make changes on the chart, save the config file. then run this function again - measures will be added
     """
     if not os.path.exists(json_path):
         if strict:

pandas_plots/pls.py

@@ -500,6 +500,7 @@ def plot_bars(
     width: int = 1600,
     title: str = None,
     use_ci: bool = False,
+    ci_agg: Literal["mean", "median"] = "mean",
     precision: int = 0,
     renderer: Literal["png", "svg", None] = "png",
     png_path: Path | str = None,
@@ -569,9 +570,9 @@ def plot_bars(
                 dropna=False,
             )
             .agg(
-                mean=(col_name, "mean"),
+                mean=(col_name, ci_agg),
                 # * retrieve margin from custom func
-                margin=(col_name, lambda x: mean_confidence_interval(x)[1]),
+                margin=(col_name, lambda x: mean_confidence_interval(x, use_median = (ci_agg == "median"))[1]),
             )
             .reset_index()
         )
@@ -653,7 +654,7 @@ def plot_bars(
 
     # * title str n
     _title_str_n = (
-        f", n={n_len:_} ({n:_})" if not use_ci else f", n={n_len:_})<br><sub>ci(95) on means<sub>"
+        f", n={n_len:_} ({n:_})" if not use_ci else f", n={n_len:_})<br><sub>ci(95) on {ci_agg}s<sub>"
     )
 
     # * title str na
@@ -965,6 +966,7 @@ def plot_box(
     violin: bool = False,
     x_min: float = None,
     x_max: float = None,
+    use_log: bool = False,
     png_path: Path | str = None,
 ) -> object:
     """
@@ -977,10 +979,13 @@ def plot_box(
         height: The height of the plot.
         width: The width of the plot.
         annotations: Whether to add annotations to the plot.
-        violin: Use violin plot or not
-        x_min: The minimum value for the x-axis scale (max and min must be set)
-        x_max: The maximum value for the x-axis scale (max and min must be set)
-        summary: Whether to add a summary table to the plot
+        summary: Whether to add a summary table to the plot.
+        caption: The caption for the plot.
+        title: The title of the plot.
+        violin: Use violin plot or not.
+        x_min: The minimum value for the x-axis scale (max and min must be set).
+        x_max: The maximum value for the x-axis scale (max and min must be set).
+        use_log: Use logarithmic scale for the axis.
         png_path (Path | str, optional): The path to save the image as a png file. Defaults to None.
 
     Returns:
@@ -993,7 +998,7 @@ def plot_box(
     # * drop na to keep scipy sane
     n_ = len(ser)
     ser.dropna(inplace=True)
-    n = len(ser)
+    # n = len(ser)
 
     # hack
     median = ser.median()
@@ -1011,7 +1016,6 @@ def plot_box(
     lvl3 = height * 0.25
 
     caption = _set_caption(caption)
-
     dict = {
         "data_frame": ser,
         "orientation": "h",
@@ -1020,7 +1024,9 @@ def plot_box(
         "width": width,
         "points": points,
         # 'box':True,
-        "title": f"{caption}[{ser.name}], n = {n_:_}({n:_})" if not title else title,
+        "log_x": use_log,   # * logarithmic scale, axis is always x
+        # "notched": True,
+        "title": f"{caption}[{ser.name}], n = {n_:_}" if not title else title,
     }
 
     fig = px.violin(**{**dict, "box": True}) if violin else px.box(**dict)
@@ -1119,6 +1125,8 @@ def plot_boxes(
     annotations: bool = True,
     summary: bool = True,
     title: str = None,
+    use_log: bool = False,
+    box_width: float = 0.5,
     png_path: Path | str = None,
 ) -> object:
     """
@@ -1133,6 +1141,7 @@ def plot_boxes(
         width (int): The width of the plot.
         annotations (bool): Whether to add annotations to the plot.
         summary (bool): Whether to add a summary to the plot.
+        use_log (bool): Whether to use logarithmic scale for the plot (cannot show negative values).
         png_path (Path | str, optional): The path to save the image as a png file. Defaults to None.
 
     Returns:
@@ -1170,11 +1179,14 @@ def plot_boxes(
         df,
         x=df.iloc[:, 0],
         y=df.iloc[:, 1],
+        color=df.iloc[:, 0],
         template="plotly_dark" if os.getenv("THEME") == "dark" else "plotly",
         orientation="v",
         height=height,
         width=width,
         points=points,
+        log_y=use_log,
+        # color_discrete_sequence=px.colors.qualitative.Plotly,
         title=(
             f"{caption}[{df.columns[0]}] on [{df.columns[1]}], n = {len(df):_.0f}"
             if not title
@@ -1245,6 +1257,9 @@ def plot_boxes(
 
     fig.update_xaxes(title_text=df.columns[0])
     fig.update_yaxes(title_text=df.columns[1])
+    fig.update_layout(boxmode="group")  # Ensures boxes are not too compressed
+    fig.update_layout(showlegend=False)
+    fig.update_traces(marker=dict(size=5), width=box_width)  # Adjust width (default ~0.5)
 
     fig.show("png")
     if summary:

pandas_plots/tbl.py

@@ -121,14 +121,14 @@ def describe_df(
         if df[col].notna().sum() == 0 and df[col].dtype == "float":
             df[col] = df[col].astype(str)
 
-    print(f"🔵 {'*'*3} df: {caption} {'*'*3}")
-    print(f"🟣 shape: ({df.shape[0]:_}, {df.shape[1]}) columns: {np.array(df.columns)} ")
+    print(f"🔵 {'*'*3} df: {caption} {'*'*3}  ")
+    print(f"🟣 shape: ({df.shape[0]:_}, {df.shape[1]}) columns: {np.array(df.columns)}  ")
     # print(f"🟣 shape: ({df.shape[0]:_}, {df.shape[1]}) columns: {df.columns.tolist()} ")
-    print(f"🟣 duplicates: {df.duplicated().sum():_}")
-    print(f"🟣 uniques: {wrap_text(str({col: f'{df[col].nunique():_}' for col in df})) }")
+    print(f"🟣 duplicates: {df.duplicated().sum():_}  ")
+    print(f"🟣 uniques: {wrap_text(str({col: f'{df[col].nunique():_}' for col in df})) }  ")
     # print(f"🟣 uniques: { {col: f'{df[col].nunique():_}' for col in df} }")
     # print(f"🟣 uniques: {{ {', '.join(f'{col}: {df[col].nunique():_}' for col in df)} }}")
-    print(f"🟣 missings: {wrap_text(str({col: f'{df[col].isna().sum():_}' for col in df})) }")
+    print(f"🟣 missings: {wrap_text(str({col: f'{df[col].isna().sum():_}' for col in df})) }  ")
     # print(f"🟣 missings: { {col: f'{df[col].isna().sum():_}' for col in df} }")
     # print(f"🟣 missings: {dict(df.isna().sum())}")
     
@@ -141,13 +141,13 @@ def describe_df(
         # unis = df[col].sort_values().unique()
         unis = list(df[col].value_counts().sort_index().index)
         # * get header
-        header = f"🟠 {col}({len(unis):_}|{df[col].dtype})"
+        header = f"🟠 {col}({len(unis):_}|{df[col].dtype})  "
         return unis, header
 
     # hack this block somehow interferes with the plotly renderer. so its run even when use_columns=False
     if use_columns:
-        print("--- column uniques (all)")
-        print(f"🟠 index {wrap_text(df.index.tolist()[:top_n_uniques])}")
+        print("--- column uniques (all)  ")
+        print(f"🟠 index {wrap_text(df.index.tolist()[:top_n_uniques])}  ")
     for col in df.columns[:]:
         _u, _h = get_uniques_header(col)
         # * check col type
@@ -155,10 +155,10 @@ def describe_df(
         # * wrap output
         if use_columns:
                 print(
-                    f"{_h} {wrap_text(_u[:top_n_uniques], max_items_in_line=70, use_apo=is_str)}"
+                    f"{_h} {wrap_text(_u[:top_n_uniques], max_items_in_line=70, use_apo=is_str)}  "
                 )
 
-    print("--- column stats (numeric)")
+    print("--- column stats (numeric)  ")
     # * only show numerics
     for col in df.select_dtypes("number").columns:
         _u, _h = get_uniques_header(col)
@@ -793,7 +793,7 @@ def print_summary(df: pd.DataFrame | pd.Series, show: bool = True, name: str=" "
         # * extra care for scipy metrics, these are very vulnarable to nan
         if show:
             print(
-                f"""{name} -> min: {min:_} | lower: {lower:_} | q25: {q1:_} | median: {med:_} | mean: {mean:_} | q75: {q3:_} | upper: {upper:_} | max: {max:_} | std: {std:_} | cv: {cv:_} | sum: {sum:_} | skew: {skew} | kurto: {kurto}""")
+                f"""{name} -> min: {min:_} | lower: {lower:_} | q25: {q1:_} | median: {med:_} | mean: {mean:_} | q75: {q3:_} | upper: {upper:_} | max: {max:_} | std: {std:_} | cv: {cv:_} | sum: {sum:_} | skew: {skew} | kurto: {kurto}  """)
 
         summary = {
             "min": min,

{pandas_plots-0.12.20.dist-info → pandas_plots-0.12.22.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: pandas-plots
-Version: 0.12.20
+Version: 0.12.22
 Summary: A collection of helper for table handling and visualization
 Home-page: https://github.com/smeisegeier/pandas-plots
 Author: smeisegeier

pandas_plots-0.12.22.dist-info/RECORD

@@ -0,0 +1,11 @@
+pandas_plots/hlp.py,sha256=i11Ep9P-u9O0bvexGTELRDUtmLzvNgNHxnkQTGf3DwQ,20838
+pandas_plots/pii.py,sha256=2WKE-W9s285jPdsTqCgt1uxuW4lj1PYCVOYB2fYDNwQ,2195
+pandas_plots/pls.py,sha256=APvF_cEYN28TtlpNNIJ2NPTA3chTP9ZHtwnVEuZ-skI,49059
+pandas_plots/tbl.py,sha256=RJWBHeKGTAhGpVCY57TsS_dYR-FpInP-TOsKW_tU4V4,32556
+pandas_plots/ven.py,sha256=2x3ACo2vSfO3q6fv-UdDQ0h1SJyt8WChBGgE5SDCdCk,11673
+pandas_plots-0.12.22.dist-info/licenses/LICENSE,sha256=6KQ5KVAAhRaB-JJKpX4cefKvRZRgI7GUPc92_2d31XY,1051
+pandas_plots-0.12.22.dist-info/METADATA,sha256=0bdvEP5M1SgmSJI3QKLd8MX1RjSrwzxlXWrygQNjHaM,7564
+pandas_plots-0.12.22.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
+pandas_plots-0.12.22.dist-info/pii.py,sha256=2WKE-W9s285jPdsTqCgt1uxuW4lj1PYCVOYB2fYDNwQ,2195
+pandas_plots-0.12.22.dist-info/top_level.txt,sha256=XnaNuIHBqMmCeh_U7nKOYTwFue_SIA0wxuDgdPmnnSk,13
+pandas_plots-0.12.22.dist-info/RECORD,,

pandas_plots-0.12.20.dist-info/RECORD

@@ -1,11 +0,0 @@
-pandas_plots/hlp.py,sha256=TxduvDztLtMuVg7rRS5mEnhktc-eHf5vI9SG1ppAbxk,20299
-pandas_plots/pii.py,sha256=2WKE-W9s285jPdsTqCgt1uxuW4lj1PYCVOYB2fYDNwQ,2195
-pandas_plots/pls.py,sha256=S9hKQTPp8XIL9RmzvfeSQE4t2jm18bG4bejwcd-dJF4,48236
-pandas_plots/tbl.py,sha256=LxMKJh4qkGuQZ1DdCZIq1tMS26F6elsqbe_uabvQx4E,32535
-pandas_plots/ven.py,sha256=2x3ACo2vSfO3q6fv-UdDQ0h1SJyt8WChBGgE5SDCdCk,11673
-pandas_plots-0.12.20.dist-info/licenses/LICENSE,sha256=6KQ5KVAAhRaB-JJKpX4cefKvRZRgI7GUPc92_2d31XY,1051
-pandas_plots-0.12.20.dist-info/METADATA,sha256=E64_SFyau96op38X5-Ld-8mVJkRr9xBKY9cJr2lu5-o,7564
-pandas_plots-0.12.20.dist-info/WHEEL,sha256=CmyFI0kx5cdEMTLiONQRbGQwjIoR1aIYB7eCAQ4KPJ0,91
-pandas_plots-0.12.20.dist-info/pii.py,sha256=2WKE-W9s285jPdsTqCgt1uxuW4lj1PYCVOYB2fYDNwQ,2195
-pandas_plots-0.12.20.dist-info/top_level.txt,sha256=XnaNuIHBqMmCeh_U7nKOYTwFue_SIA0wxuDgdPmnnSk,13
-pandas_plots-0.12.20.dist-info/RECORD,,