dragon-ml-toolbox 19.12.2__py3-none-any.whl → 19.14.0__py3-none-any.whl

{dragon_ml_toolbox-19.12.2.dist-info → dragon_ml_toolbox-19.14.0.dist-info}/METADATA

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: dragon-ml-toolbox
-Version: 19.12.2
+Version: 19.14.0
 Summary: Complete pipelines and helper tools for data science and machine learning projects.
 Author-email: Karl Luigi Loza Vidaurre <luigiloza@gmail.com>
 License-Expression: MIT

{dragon_ml_toolbox-19.12.2.dist-info → dragon_ml_toolbox-19.14.0.dist-info}/RECORD

@@ -1,5 +1,5 @@
-dragon_ml_toolbox-19.12.2.dist-info/licenses/LICENSE,sha256=L35WDmmLZNTlJvxF6Vy7Uy4SYNi6rCfWUqlTHpoRMoU,1081
-dragon_ml_toolbox-19.12.2.dist-info/licenses/LICENSE-THIRD-PARTY.md,sha256=0-HBRMMgKuwtGy6nMJZvIn1fLxhx_ksyyVB2U_iyYZU,2818
+dragon_ml_toolbox-19.14.0.dist-info/licenses/LICENSE,sha256=L35WDmmLZNTlJvxF6Vy7Uy4SYNi6rCfWUqlTHpoRMoU,1081
+dragon_ml_toolbox-19.14.0.dist-info/licenses/LICENSE-THIRD-PARTY.md,sha256=0-HBRMMgKuwtGy6nMJZvIn1fLxhx_ksyyVB2U_iyYZU,2818
 ml_tools/ETL_cleaning.py,sha256=cKXyRFaaFs_beAGDnQM54xnML671kq-yJEGjHafW-20,351
 ml_tools/ETL_engineering.py,sha256=cwh1FhtNdUHllUDvho-x3SIVj4KwG_rFQR6VYzWUg0U,898
 ml_tools/GUI_tools.py,sha256=O89rG8WQv6GY1DiphQjIsPzXFCQID6te7q_Sgt1iTkQ,294
@@ -88,7 +88,7 @@ ml_tools/_core/_PSO_optimization.py,sha256=W3g5xw2v2eOUQadv8KHFkt5HNm9AiY3ZUk-Te
 ml_tools/_core/_SQL.py,sha256=zX_8EgYfmLmvvrnL851KMkI4w9kdkjHJ997BTvS5aig,11556
 ml_tools/_core/_VIF_factor.py,sha256=BM0mTowBqt45PXFy9oJLhT9C-CTWWo0TQhgCyWYLHtQ,10457
 ml_tools/_core/__init__.py,sha256=d4IG0OxUXj2HffepzQcYixHlZeuuuDMAFa09H_6LtmU,12
-ml_tools/_core/_data_exploration.py,sha256=uynIjMppbr5nFJ-7wag0R0HDQCp2rTXqz_IpgNKKNOM,77508
+ml_tools/_core/_data_exploration.py,sha256=tOdtXTCh_xESKqIUuxCCo8fbcPoO9Eu5PwJwyehGKY8,76434
 ml_tools/_core/_ensemble_evaluation.py,sha256=17lWl4bWLT1BAMv_fhGf2D3wy-F4jx0HgnJ79lYkRuE,28419
 ml_tools/_core/_ensemble_inference.py,sha256=9UpARSETzmqPdQmxqizD768tjkqldxHw1ER_hM9Kx9M,8631
 ml_tools/_core/_ensemble_learning.py,sha256=X8ghbjDOLMENCWdISXLhDlHQtR3C6SW1tkTBAcfRRPY,22016
@@ -105,7 +105,7 @@ ml_tools/_core/_schema.py,sha256=TM5WVVMoKOvr_Bc2z34sU_gzKlM465PRKTgdZaEOkGY,140
 ml_tools/_core/_script_info.py,sha256=21r83LV3RubsNZ_RTEUON6RbDf7Mh4_udweNcvdF_Fk,212
 ml_tools/_core/_serde.py,sha256=tsI4EO2Y7jrBMmbQ1pinDsPOrOg-SaPuB-Dt40q0taE,5609
 ml_tools/_core/_utilities.py,sha256=oU-0hBipE96bXox66NG-hFuEMMNkKa9MkAy1yJGCSIA,22779
-dragon_ml_toolbox-19.12.2.dist-info/METADATA,sha256=Nuk7YVRdDotD_TURCpIFKqcDuTlkBs9fpTv-8jCm5aU,8193
-dragon_ml_toolbox-19.12.2.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
-dragon_ml_toolbox-19.12.2.dist-info/top_level.txt,sha256=wm-oxax3ciyez6VoO4zsFd-gSok2VipYXnbg3TH9PtU,9
-dragon_ml_toolbox-19.12.2.dist-info/RECORD,,
+dragon_ml_toolbox-19.14.0.dist-info/METADATA,sha256=7QaJsWeT9idUhpAV37t64fsuWNmzXbaQqA-a-yDP2yY,8193
+dragon_ml_toolbox-19.14.0.dist-info/WHEEL,sha256=_zCd3N1l69ArxyTb8rzEoP9TpbYXkqRFSNOD5OuxnTs,91
+dragon_ml_toolbox-19.14.0.dist-info/top_level.txt,sha256=wm-oxax3ciyez6VoO4zsFd-gSok2VipYXnbg3TH9PtU,9
+dragon_ml_toolbox-19.14.0.dist-info/RECORD,,

ml_tools/_core/_data_exploration.py

@@ -26,21 +26,21 @@ __all__ = [
     "drop_macro",
     "clean_column_names",
     "plot_value_distributions", 
-    "plot_continuous_vs_target",
-    "plot_categorical_vs_target",
-    "encode_categorical_features",
     "split_features_targets", 
-    "split_continuous_binary", 
+    "encode_categorical_features",
     "clip_outliers_single", 
     "clip_outliers_multi",
     "drop_outlier_samples",
+    "plot_continuous_vs_target",
+    "plot_categorical_vs_target",
     "plot_correlation_heatmap", 
+    "finalize_feature_schema",
     "match_and_filter_columns_by_regex",
     "standardize_percentages",
     "reconstruct_one_hot",
     "reconstruct_binary",
     "reconstruct_multibinary",
-    "finalize_feature_schema",
+    "split_continuous_binary", 
     "apply_feature_schema"
 ]
 
@@ -59,16 +59,18 @@ def summarize_dataframe(df: pd.DataFrame, round_digits: int = 2):
     """
     summary = pd.DataFrame({
         'Data Type': df.dtypes,
-        'Non-Null Count': df.notnull().sum(),
+        'Completeness %': (df.notnull().mean() * 100).round(2),
         'Unique Values': df.nunique(),
-        'Missing %': (df.isnull().mean() * 100).round(round_digits)
+        # 'Missing %': (df.isnull().mean() * 100).round(2)
     })
 
     # For numeric columns, add summary statistics
     numeric_cols = df.select_dtypes(include='number').columns
     if not numeric_cols.empty:
-        summary_numeric = df[numeric_cols].describe().T[
-            ['mean', 'std', 'min', '25%', '50%', '75%', 'max']
+        stats = df[numeric_cols].describe(percentiles=[.10, .25, .50, .70, .80, .90])
+        
+        summary_numeric = stats.T[
+            ['mean', 'std', 'min', '10%', '25%', '50%', '70%', '80%', '90%', 'max']
         ].round(round_digits)
         summary = summary.join(summary_numeric, how='left')
 
@@ -108,22 +110,17 @@ def drop_constant_columns(df: pd.DataFrame, verbose: bool = True) -> pd.DataFram
     for col_name in df_clean.columns:
         column = df_clean[col_name]
         
-        # We can apply this logic to all columns or only focus on numeric ones.
-        # if not is_numeric_dtype(column):
-        #     cols_to_keep.append(col_name)
-        #     continue
-        
         # Keep a column if it has more than one unique value (nunique ignores NaNs by default)
         if column.nunique(dropna=True) > 1:
             cols_to_keep.append(col_name)
 
     dropped_columns = original_columns - set(cols_to_keep)
     if verbose:
-        _LOGGER.info(f"🧹 Dropped {len(dropped_columns)} constant columns.")
         if dropped_columns:
-            for dropped_column in dropped_columns:
-                print(f"    {dropped_column}")
-                
+            _LOGGER.info(f"🧹 Dropped {len(dropped_columns)} constant columns: {list(dropped_columns)}")
+        else:
+            _LOGGER.info("No constant columns found.")
+            
     # Return a new DataFrame with only the columns to keep
     df_clean = df_clean[cols_to_keep]
     
@@ -338,8 +335,7 @@ def drop_columns_with_missing_data(df: pd.DataFrame, threshold: float = 0.7, sho
     cols_to_drop = missing_fraction[missing_fraction > threshold].index # type: ignore
 
     if len(cols_to_drop) > 0:
-        _LOGGER.info(f"🧹 Dropping columns with more than {threshold*100:.0f}% missing data:")
-        print(list(cols_to_drop))
+        _LOGGER.info(f"🧹 Dropping columns with more than {threshold*100:.0f}% missing data: {list(cols_to_drop)}")
         
         result_df = df.drop(columns=cols_to_drop)
         if show_nulls_after:
@@ -369,9 +365,8 @@ def drop_macro(df: pd.DataFrame,
 
     Args:
         df (pd.DataFrame): The input pandas DataFrame to be cleaned.
-        log_directory (Union[str, Path]): Path to the directory where the
-            'Missing_Data_start.csv' and 'Missing_Data_final.csv' logs
-            will be saved.
+        log_directory (Union[str, Path]): Path to the directory where the missing data reports
+            and plots will be saved inside a "Missing Report" subdirectory.
         targets (list[str]): A list of column names to be treated as target
             variables. This list guides the row-dropping logic.
         skip_targets (bool, optional): If True, the columns listed in `targets`
@@ -387,15 +382,18 @@ def drop_macro(df: pd.DataFrame,
     # make a deep copy to work with
     df_clean = df.copy()
     
+    base_dir_path = make_fullpath(log_directory, make=True, enforce="directory")
+    full_path = base_dir_path / "Missing Report"
+    
     # Log initial state + Plot
     missing_data_start = show_null_columns(
         df=df_clean, 
-        plot_to_dir=log_directory, 
+        plot_to_dir=full_path, 
         plot_filename="Original",
         use_all_columns=True
     )
     save_dataframe_filename(df=missing_data_start.reset_index(drop=False),
-                   save_dir=log_directory,
+                   save_dir=full_path,
                    filename="Missing_Data_Original")
     
     # Clean cycles for rows and columns
@@ -424,12 +422,12 @@ def drop_macro(df: pd.DataFrame,
     # log final state + plot
     missing_data_final = show_null_columns(
         df=df_clean,
-        plot_to_dir=log_directory,
+        plot_to_dir=full_path,
         plot_filename="Processed",
         use_all_columns=True
     )
     save_dataframe_filename(df=missing_data_final.reset_index(drop=False),
-                   save_dir=log_directory,
+                   save_dir=full_path,
                    filename="Missing_Data_Processed")
     
     # return cleaned dataframe
@@ -476,9 +474,8 @@ def plot_value_distributions(
     df: pd.DataFrame,
     save_dir: Union[str, Path],
     categorical_columns: Optional[List[str]] = None,
-    categorical_cardinality_threshold: int = 10,
-    max_categories: int = 50,
-    fill_na_with: str = "Missing"
+    max_categories: int = 100,
+    fill_na_with: str = "MISSING DATA"
 ):
     """
     Plots and saves the value distributions for all columns in a DataFrame,
@@ -491,15 +488,9 @@ def plot_value_distributions(
     Args:
         df (pd.DataFrame): The input DataFrame to analyze.
         save_dir (str | Path): Directory path to save the plots.
-        categorical_columns (List[str] | None): If provided, this list
-            of column names will be treated as categorical, and all other columns will be treated as continuous. This
-            overrides the `continuous_cardinality_threshold` logic.
-        categorical_cardinality_threshold (int): A numeric column will be treated
-            as 'categorical' if its number of unique values is less than or equal to this threshold. (Ignored if `categorical_columns` is set).
-        max_categories (int): The maximum number of unique categories a
-            categorical feature can have to be plotted. Features exceeding this limit will be skipped.
-        fill_na_with (str): A string to replace NaN values in categorical columns. This allows plotting 'missingness' as its
-            own category. Defaults to "Missing".
+        categorical_columns (List[str] | None): If provided, these will be treated as categorical, and all other columns will be treated as continuous.
+        max_categories (int): The maximum number of unique categories a categorical feature can have to be plotted. Features exceeding this limit will be skipped.
+        fill_na_with (str): A string to replace NaN values in categorical columns. This allows plotting 'missingness' as its own category.
 
     Notes:
         - `seaborn.histplot` with KDE is used for continuous features.
@@ -534,7 +525,7 @@ def plot_value_distributions(
                     is_continuous = True
             else:
                 # Use auto-detection
-                if is_numeric and n_unique > categorical_cardinality_threshold:
+                if is_numeric:
                     is_continuous = True
             
             # --- Case 1: Continuous Numeric (Histogram) ---
@@ -549,7 +540,7 @@ def plot_value_distributions(
                 save_path = numeric_dir / f"{sanitize_filename(col_name)}.svg"
                 numeric_plots_saved += 1
 
-            # --- Case 2: Categorical or Low-Cardinality Numeric (Count Plot) ---
+            # --- Case 2: Categorical (Count Plot) ---
             else:
                 # Check max categories
                 if n_unique > max_categories:
@@ -558,7 +549,7 @@ def plot_value_distributions(
 
                 # Adaptive figure size
                 fig_width = max(10, n_unique * 0.5)
-                plt.figure(figsize=(fig_width, 7))
+                plt.figure(figsize=(fig_width, 8))
                 
                 # Make a temporary copy for plotting to handle NaNs
                 temp_series = df[col_name].copy()
@@ -573,7 +564,7 @@ def plot_value_distributions(
                 
                 # Get category order by frequency
                 order = temp_series.value_counts().index
-                sns.countplot(x=temp_series, order=order, palette="viridis")
+                sns.countplot(x=temp_series, order=order, palette="Oranges", hue=temp_series, legend=False)
                 
                 plt.title(f"Distribution of '{col_name}' (Categorical)")
                 plt.xlabel(col_name)
@@ -607,68 +598,55 @@ def plot_value_distributions(
 
 
 def plot_continuous_vs_target(
-    df: pd.DataFrame,
-    targets: List[str],
+    df_continuous: pd.DataFrame,
+    df_targets: pd.DataFrame,
     save_dir: Union[str, Path],
-    features: Optional[List[str]] = None
+    verbose: int = 1
 ):
     """
-    Plots each continuous feature against each target to visualize linear relationships.
+    Plots each continuous feature from df_continuous against each target in df_targets.
 
-    This function is a common EDA step for regression tasks. It creates a
-    scatter plot for each feature-target pair, overlays a simple linear
-    regression line, and saves each plot as an individual .svg file.
+    This function creates a scatter plot for each feature-target pair, overlays a 
+    simple linear regression line, and saves each plot as an individual .svg file.
 
     Plots are saved in a structured way, with a subdirectory created for
     each target variable.
 
     Args:
-        df (pd.DataFrame): The input DataFrame.
-        targets (List[str]): A list of target column names to plot (y-axis).
-        save_dir (str | Path): The base directory where plots will be saved. A subdirectory will be created here for each target.
-        features (List[str] | None): A list of feature column names to plot (x-axis). If None, all non-target columns in the
-            DataFrame will be used.
+        df_continuous (pd.DataFrame): DataFrame containing continuous feature columns (x-axis).
+        df_targets (pd.DataFrame): DataFrame containing target columns (y-axis).
+        save_dir (str | Path): The base directory where plots will be saved.
+        verbose (int): Verbosity level for logging warnings.
 
     Notes:
-        - Only numeric features and numeric targets are processed. Non-numeric
-          columns in the lists will be skipped with a warning.
-        - Rows with NaN in either the feature or the target are dropped
-          pairwise for each plot.
+        - Only numeric features and numeric targets are processed.
+        - Rows with NaN in either the feature or the target are dropped pairwise.
+        - Assumes df_continuous and df_targets share the same index.
     """
     # 1. Validate the base save directory
     base_save_path = make_fullpath(save_dir, make=True, enforce="directory")
 
-    # 2. Validate helper
-    def _validate_numeric_cols(col_list: List[str], col_type: str) -> List[str]:
+    # 2. Validation helper
+    def _get_valid_numeric_cols(df: pd.DataFrame, df_name: str) -> List[str]:
         valid_cols = []
-        for col in col_list:
-            if col not in df.columns:
-                _LOGGER.warning(f"{col_type} column '{col}' not found. Skipping.")
-            elif not is_numeric_dtype(df[col]):
-                _LOGGER.warning(f"{col_type} column '{col}' is not numeric. Skipping.")
+        for col in df.columns:
+            if not is_numeric_dtype(df[col]):
+                if verbose > 0:
+                    _LOGGER.warning(f"Column '{col}' in {df_name} is not numeric. Skipping.")
             else:
                 valid_cols.append(col)
         return valid_cols
 
-    # 3. Validate target columns FIRST
-    valid_targets = _validate_numeric_cols(targets, "Target")
+    # 3. Validate target columns
+    valid_targets = _get_valid_numeric_cols(df_targets, "df_targets")
     if not valid_targets:
-        _LOGGER.error("No valid numeric target columns provided to plot.")
+        _LOGGER.error("No valid numeric target columns provided in df_targets.")
         return
     
-    # 4. Determine and validate feature columns
-    if features is None:
-        _LOGGER.info("No 'features' list provided. Using all non-target columns as features.")
-        target_set = set(valid_targets)
-        # Get all columns that are not in the valid_targets set
-        features_to_validate = [col for col in df.columns if col not in target_set]
-    else:
-        features_to_validate = features
-        
-    valid_features = _validate_numeric_cols(features_to_validate, "Feature")
-
+    # 4. Validate feature columns
+    valid_features = _get_valid_numeric_cols(df_continuous, "df_continuous")
     if not valid_features:
-        _LOGGER.error("No valid numeric feature columns found to plot.")
+        _LOGGER.error("No valid numeric feature columns provided in df_continuous.")
         return
 
     # 5. Main plotting loop
@@ -680,15 +658,20 @@ def plot_continuous_vs_target(
         target_save_dir = base_save_path / safe_target_dir_name
         target_save_dir.mkdir(parents=True, exist_ok=True)
         
-        _LOGGER.info(f"Generating plots for target: '{target_name}' -> Saving to '{target_save_dir.name}'")
+        if verbose > 0:
+            _LOGGER.info(f"Generating plots for target: '{target_name}' -> Saving to '{target_save_dir.name}'")
 
         for feature_name in valid_features:
             
-            # Drop NaNs pairwise for this specific plot
-            temp_df = df[[feature_name, target_name]].dropna()
+            # Align data and drop NaNs pairwise - use concat to ensure we respect the index alignment between the two DFs
+            temp_df = pd.concat([
+                df_continuous[feature_name], 
+                df_targets[target_name]
+            ], axis=1).dropna()
 
             if temp_df.empty:
-                _LOGGER.warning(f"No non-null data for '{feature_name}' vs '{target_name}'. Skipping plot.")
+                if verbose > 1:
+                    _LOGGER.warning(f"No non-null data for '{feature_name}' vs '{target_name}'. Skipping plot.")
                 continue
 
             x = temp_df[feature_name]
@@ -696,11 +679,12 @@ def plot_continuous_vs_target(
 
             # 6. Perform linear fit
             try:
-                # Modern replacement for np.polyfit + np.poly1d. Compatible with NumPy 1.14+ and NumPy 2.0+
+                # Modern replacement for np.polyfit + np.poly1d
                 p = np.polynomial.Polynomial.fit(x, y, deg=1)
                 plot_regression_line = True
             except (np.linalg.LinAlgError, ValueError):
-                _LOGGER.warning(f"Linear regression failed for '{feature_name}' vs '{target_name}'. Plotting scatter only.")
+                if verbose > 0:
+                    _LOGGER.warning(f"Linear regression failed for '{feature_name}' vs '{target_name}'. Plotting scatter only.")
                 plot_regression_line = False
 
             # 7. Create the plot
@@ -734,87 +718,68 @@ def plot_continuous_vs_target(
             
             # Close the figure to free up memory
             plt.close()
-
-    _LOGGER.info(f"Successfully saved {total_plots_saved} feature-vs-target plots to '{base_save_path}'.")
+    
+    if verbose > 0:
+        _LOGGER.info(f"Successfully saved {total_plots_saved} feature-vs-target plots to '{base_save_path}'.")
 
 
 def plot_categorical_vs_target(
-    df: pd.DataFrame,
-    targets: List[str],
+    df_categorical: pd.DataFrame,
+    df_targets: pd.DataFrame,
     save_dir: Union[str, Path],
-    features: Optional[List[str]] = None,
-    plot_type: Literal["box", "violin"] = "box",
-    max_categories: int = 20,
-    fill_na_with: str = "Missing"
+    max_categories: int = 50,
+    fill_na_with: str = "MISSING DATA",
+    drop_empty_targets: bool = True,
+    verbose: int = 1
 ):
     """
-    Plots each categorical feature against each numeric target using box or violin plots.
+    Plots each feature in df_categorical against each numeric target in df_targets using box plots.
 
-    This function is a core EDA step for regression tasks to understand the
-    relationship between a categorical independent variable and a continuous
-    dependent variable.
-    
-    Plots are saved as individual .svg files in a structured way, with a subdirectory created for each target.
+    Automatically aligns the two DataFrames by index. If a numeric
+    column is passed within df_categorical, it will be cast to object type to treat it as a category.
 
     Args:
-        df (pd.DataFrame): The input DataFrame.
-        targets (List[str]): A list of numeric target column names (y-axis).
-        save_dir (str | Path): The base directory where plots will be saved. A subdirectory will be created here for each target.
-        features (List[str] | None): A list of categorical feature column names (x-axis). If None, all non-numeric (object) columns will be used.
-        plot_type (Literal["box", "violin"]): The type of plot to generate.
-        max_categories (int): The maximum number of unique categories a feature can have to be plotted. Features exceeding this limit will be skipped.
-        fill_na_with (str): A string to replace NaN values in categorical columns. This allows plotting 'missingness' as its own category. Defaults to "Missing".
+        df_categorical (pd.DataFrame): DataFrame containing categorical feature columns (x-axis).
+        df_targets (pd.DataFrame): DataFrame containing numeric target columns (y-axis).
+        save_dir (str | Path): Base directory for saving plots.
+        max_categories (int): The maximum number of unique categories a feature can have to be plotted.
+        fill_na_with (str): String to replace NaN values in categorical columns.
+        drop_empty_targets (bool): If True, drops rows where the target value is NaN before plotting.
+        verbose (int): Verbosity level for logging warnings.
 
     Notes:
-        - Only numeric targets are processed.
-        - Features are automatically identified as categorical if they are 'object' dtype.
+        - Assumes df_categorical and df_targets share the same index.
     """
-    # 1. Validate the base save directory and inputs
+    # 1. Validate the base save directory
     base_save_path = make_fullpath(save_dir, make=True, enforce="directory")
-    
-    if plot_type not in ["box", "violin"]:
-        _LOGGER.error(f"Invalid plot type '{plot_type}'")
-        raise ValueError()
 
     # 2. Validate target columns (must be numeric)
     valid_targets = []
-    for col in targets:
-        if col not in df.columns:
-            _LOGGER.warning(f"Target column '{col}' not found. Skipping.")
-        elif not is_numeric_dtype(df[col]):
-            _LOGGER.warning(f"Target column '{col}' is not numeric. Skipping.")
+    for col in df_targets.columns:
+        if not is_numeric_dtype(df_targets[col]):
+            if verbose > 0:
+                _LOGGER.warning(f"Target column '{col}' in df_targets is not numeric. Skipping.")
         else:
             valid_targets.append(col)
     
     if not valid_targets:
-        _LOGGER.error("No valid numeric target columns provided to plot.")
+        _LOGGER.error("No valid numeric target columns provided in df_targets.")
         return
 
-    # 3. Determine and validate feature columns
-    features_to_plot = []
-    if features is None:
-        _LOGGER.info("No 'features' list provided. Auto-detecting categorical features.")
-        for col in df.columns:
-            if col in valid_targets:
-                continue
-            
-            # Auto-include object dtypes
-            if is_object_dtype(df[col]):
-                features_to_plot.append(col)
-            # Auto-include low-cardinality numeric features - REMOVED
-            # elif is_numeric_dtype(df[col]) and df[col].nunique() <= max_categories:
-            #     _LOGGER.info(f"Treating low-cardinality numeric column '{col}' as categorical.")
-            #     features_to_plot.append(col)
-    else:
-        # Validate user-provided list
-        for col in features:
-            if col not in df.columns:
-                _LOGGER.warning(f"Feature column '{col}' not found in DataFrame. Skipping.")
-            else:
-                features_to_plot.append(col)
+    # 3. Validate feature columns (Flexible: Allow numeric but warn)
+    valid_features = []
+    for col in df_categorical.columns:
+        # If numeric, warn but accept it (will be cast to object later)
+        if is_numeric_dtype(df_categorical[col]):
+            if verbose > 0:
+                _LOGGER.warning(f"Feature '{col}' in df_categorical is numeric. It will be cast to 'object' and treated as categorical.")
+            valid_features.append(col)
+        else:
+            # Assume it is already object/category
+            valid_features.append(col)
 
-    if not features_to_plot:
-        _LOGGER.error("No valid categorical feature columns found to plot.")
+    if not valid_features:
+        _LOGGER.error("No valid feature columns provided in df_categorical.")
         return
 
     # 4. Main plotting loop
@@ -822,39 +787,53 @@ def plot_categorical_vs_target(
     
     for target_name in valid_targets:
         # Create a sanitized subdirectory for this target
-        safe_target_dir_name = sanitize_filename(f"{target_name}_vs_Categorical_{plot_type}")
+        safe_target_dir_name = sanitize_filename(f"{target_name}_vs_Categorical")
         target_save_dir = base_save_path / safe_target_dir_name
         target_save_dir.mkdir(parents=True, exist_ok=True)
         
-        _LOGGER.info(f"Generating '{plot_type}' plots for target: '{target_name}' -> Saving to '{target_save_dir.name}'")
+        if verbose > 0:
+            _LOGGER.info(f"Generating plots for target: '{target_name}' -> Saving to '{target_save_dir.name}'")
+        
+        for feature_name in valid_features:
+            
+            # Align data using concat to respect indices
+            feature_series = df_categorical[feature_name]
+            target_series = df_targets[target_name]
+
+            # Create a temporary DataFrame for this pair
+            temp_df = pd.concat([feature_series, target_series], axis=1)
+
+            # Optional: Drop rows where the target is NaN
+            if drop_empty_targets:
+                temp_df = temp_df.dropna(subset=[target_name])
+                if temp_df.empty:
+                    if verbose > 1:
+                        _LOGGER.warning(f"No valid data left for '{feature_name}' vs '{target_name}' after dropping empty targets. Skipping.")
+                    continue
 
-        for feature_name in features_to_plot:
+            # Force feature to object if it isn't already (handling the numeric flexibility)
+            if not is_object_dtype(temp_df[feature_name]):
+                temp_df[feature_name] = temp_df[feature_name].astype(object)
+
+            # Handle NaNs in the feature column (treat as a category)
+            if temp_df[feature_name].isnull().any():
+                temp_df[feature_name] = temp_df[feature_name].fillna(fill_na_with)
             
-            # Make a temporary copy for plotting to handle NaNs and dtypes
-            temp_df = df[[feature_name, target_name]].copy()
+            # Convert to string to ensure consistent plotting and cardinality check
+            temp_df[feature_name] = temp_df[feature_name].astype(str)
 
             # Check cardinality
             n_unique = temp_df[feature_name].nunique()
             if n_unique > max_categories:
-                _LOGGER.warning(f"Skipping '{feature_name}': {n_unique} unique values > {max_categories} max_categories.")
+                if verbose > 1:
+                    _LOGGER.warning(f"Skipping '{feature_name}': {n_unique} unique categories > {max_categories} max_categories.")
                 continue
-            
-            # Handle NaNs by replacing them with the specified string
-            if temp_df[feature_name].isnull().any():
-                # Convert to object type first to allow string replacement
-                temp_df[feature_name] = temp_df[feature_name].astype(object).fillna(fill_na_with)
-            
-            # Convert feature to string to ensure correct plotting order
-            temp_df[feature_name] = temp_df[feature_name].astype(str)
 
             # 5. Create the plot
-            # Increase figure width for categories
-            plt.figure(figsize=(max(10, n_unique * 1.2), 7))
+            # Dynamic figure width based on number of categories
+            plt.figure(figsize=(max(10, n_unique * 0.8), 10))
             
-            if plot_type == "box":
-                sns.boxplot(x=feature_name, y=target_name, data=temp_df)
-            elif plot_type == "violin":
-                sns.violinplot(x=feature_name, y=target_name, data=temp_df)
+            sns.boxplot(x=feature_name, y=target_name, data=temp_df)
 
             plt.title(f'{target_name} vs {feature_name}')
             plt.xlabel(feature_name)
@@ -875,8 +854,9 @@ def plot_categorical_vs_target(
                 _LOGGER.error(f"Failed to save plot: {plot_path}. Error: {e}")
             
             plt.close()
-
-    _LOGGER.info(f"Successfully saved {total_plots_saved} categorical-vs-target plots to '{base_save_path}'.")
+    
+    if verbose > 0:
+        _LOGGER.info(f"Successfully saved {total_plots_saved} categorical-vs-target plots to '{base_save_path}'.")
 
 
 def encode_categorical_features(
@@ -982,7 +962,7 @@ def encode_categorical_features(
 
     # Handle the dataset splitting logic
     if split_resulting_dataset:
-        df_categorical = df_encoded[valid_columns].to_frame() # type: ignore
+        df_categorical = df_encoded[valid_columns]
         df_non_categorical = df.drop(columns=valid_columns)
         return mappings, df_non_categorical, df_categorical
     else:
@@ -1103,7 +1083,10 @@ def plot_correlation_heatmap(df: pd.DataFrame,
         annot=annot_bool,
         cmap='coolwarm',
         fmt=".2f",
-        cbar_kws={"shrink": 0.8}
+        cbar_kws={"shrink": 0.8},
+        vmin=-1,  # Anchors minimum color to -1
+        vmax=1,   # Anchors maximum color to 1
+        center=0  # Ensures 0 corresponds to the neutral color (white)
     )
     
     # add suffix to title