Added Pandas Create from Tensor, Pandas Rename Advanced nodes. Added an option to reindex the column in Pandas Horizontal Concat node.

modules/matplotlib/mpl_bar.py

@@ -4,6 +4,7 @@ from io import StringIO
 import pandas as pd
 from .common import common_input_types
 from .util import plot_post_steps
+from ..pandas_wrapper.utils import column_label_string_to_target_type
 
 class MPLBar:
     """
@@ -56,6 +57,8 @@ class MPLBar:
 
         # Create the plot
         fig, ax = plt.subplots()
+        x_column_name = column_label_string_to_target_type(df, x_column_name)
+        y_column_name = column_label_string_to_target_type(df, y_column_name)
         ax.bar(df[x_column_name], df[y_column_name])
 
         return plot_post_steps(

modules/matplotlib/mpl_line.py

@@ -4,6 +4,7 @@ from io import StringIO
 import pandas as pd
 from .common import common_input_types
 from .util import plot_post_steps
+from ..pandas_wrapper.utils import column_label_string_to_target_type
 
 class MPLLine:
     """
@@ -53,6 +54,8 @@ class MPLLine:
         """
 
         df = dataframe
+        x_column_name = column_label_string_to_target_type(df, x_column_name)
+        y_column_name = column_label_string_to_target_type(df, y_column_name)
 
         # Create the plot
         fig, ax = plt.subplots()

modules/matplotlib/mpl_pie_chart.py

@@ -2,6 +2,7 @@ from typing import Any, Dict
 from matplotlib import pyplot as plt
 import pandas as pd
 from .util import plot_post_steps
+from ..pandas_wrapper.utils import column_label_string_to_target_type
 
 class MPLPieChart:
     """
@@ -54,6 +55,8 @@ class MPLPieChart:
         """
 
         df = dataframe
+        label_column_name = column_label_string_to_target_type(df, label_column_name)
+        data_column_name = column_label_string_to_target_type(df, data_column_name)
 
         # Create the plot
         fig, ax = plt.subplots()

modules/matplotlib/mpl_scatter.py

@@ -4,6 +4,7 @@ from io import StringIO
 import pandas as pd
 from .common import common_input_types
 from .util import plot_post_steps
+from ..pandas_wrapper.utils import column_label_string_to_target_type
 
 class MPLScatter:
     """
@@ -53,6 +54,8 @@ class MPLScatter:
         """
 
         df = dataframe
+        x_column_name = column_label_string_to_target_type(df, x_column_name)
+        y_column_name = column_label_string_to_target_type(df, y_column_name)
 
         # Create the plot
         fig, ax = plt.subplots()

modules/pandas_wrapper/pandas_create_from_tensor.py

@@ -0,0 +1,48 @@
+from typing import Any, Dict
+import pandas as pd
+import torch
+
+
+class PandasCreateFromTensor:
+    """
+    Creates a pandas DataFrame from a PyTorch tensor.
+
+    category: IO
+    """
+    
+    @classmethod
+    def INPUT_TYPES(cls) -> Dict[str, Any]:
+        """
+        Defines the input types for the `create` function.
+
+        Returns:
+            Dict[str, Any]: A dictionary specifying required input types.
+        """
+        return {
+            "required": {
+                "tens": ("TENSOR", {}),
+            }
+        }
+
+    RETURN_TYPES: tuple = ("DATAFRAME",)
+    FUNCTION: str = "f"
+    CATEGORY: str = "Data Analysis"
+
+    def f(self, tens: torch.Tensor) -> tuple:
+        """
+        Creates a pandas DataFrame from a PyTorch tensor.
+
+        Args:
+            tens (torch.Tensor): A PyTorch tensor.
+
+        Returns:
+            tuple: A tuple containing the DataFrame.
+        
+        Raises:
+            ValueError: If the tensor is not rank 1 or rank 2.
+        """
+        if tens.dim() not in [1, 2]:
+            raise ValueError("Only rank 1 and rank 2 tensors are supported.")
+
+        return (pd.DataFrame(tens.numpy()),)
+

modules/pandas_wrapper/pandas_horizontal_concat.py

@@ -5,6 +5,8 @@ import pandas as pd
 class PandasHorizontalConcat:
     """
     Horizontally concatenates two pandas DataFrames.
+    
+    If reset labels is set to True, column labels are reset to a sequential integer index.
 
     category: Transformation
     """
@@ -20,7 +22,8 @@ class PandasHorizontalConcat:
         return {
             "required": {
                 "a_dataframe": ("DATAFRAME", {}),
-                "b_dataframe": ("DATAFRAME", {})
+                "b_dataframe": ("DATAFRAME", {}),
+                "reset_labels": ("BOOLEAN", {"default": False})
             }
         }
 
@@ -28,16 +31,23 @@ class PandasHorizontalConcat:
     FUNCTION: str = "f"
     CATEGORY: str = "Data Analysis"
 
-    def f(self, a_dataframe: pd.DataFrame, b_dataframe: pd.DataFrame) -> tuple:
+    def f(self,
+          a_dataframe: pd.DataFrame,
+          b_dataframe: pd.DataFrame,
+          reset_labels: bool) -> tuple:
         """
         Horizontally concatenate two pandas DataFrames.
 
         Args:
             a_dataframe (DataFrame): The first DataFrame.
             b_dataframe (DataFrame): The second DataFrame.
+            reset_labels (bool): If reset labels is set to True, column labels are reset to a sequential integer index.
 
         Returns:
             tuple: A tuple containing the concatenated DataFrame.
         """
         df_out = pd.concat([a_dataframe, b_dataframe], axis=1)
+        if reset_labels:
+            df_out.columns = range(df_out.shape[1])
+
         return (df_out,)

modules/pandas_wrapper/pandas_rename_advanced.py

@@ -0,0 +1,72 @@
+from typing import Any, Dict
+import pandas as pd
+import ast
+
+class PandasRenameAdvanced:
+    """
+    Renames one or more indices (row labels) or column labels to new labels.
+
+    Specify the replacement in a Python dictionary format:
+    ```
+    {
+        "old_foo": "new_bar",
+        0: 5
+    }
+    ```
+    A string label has to be double-quoted.
+
+    category: Transformation
+    """
+
+    @classmethod
+    def INPUT_TYPES(cls) -> Dict[str, Any]:
+        """
+        Defines the input types for the function.
+
+        Returns:
+            Dict[str, Any]: A dictionary specifying required input types.
+        """
+        return {
+            "required": {
+                "dataframe": ("DATAFRAME", {}),
+                "axis": (("columns", "index"),),
+                "replacement_pairs": ("STRING", {"default":"{}", "multiline": True})
+            }
+        }
+
+    RETURN_TYPES: tuple = ("DATAFRAME",)
+    FUNCTION: str = "f"
+    CATEGORY: str = "Data Analysis"
+
+    def f(self, dataframe: pd.DataFrame, axis: str, replacement_pairs: str) -> tuple:
+        """
+        Renames one or more indices (row labels) or column labels to new labels.
+
+        Args:
+            dataframe (DataFrame): The DataFrame.
+            axis (str): Index (row label) or column label.
+            replacement_pairs (str): Replacement pairs specified in the Python dictionary format.
+
+        Returns:
+            tuple: A tuple containing the DataFrame.
+        """
+        try:
+            d = ast.literal_eval(replacement_pairs)
+            if not isinstance(d, dict):
+                raise ValueError("replacement_pairs must be a dictionary.")
+
+            # Check that values in the dictionary are not lists, tuples, or dictionaries
+            if any(isinstance(value, (list, tuple, dict)) for value in d.values()):
+                raise ValueError("replacement_pairs values must not be lists, tuples, or dictionaries.")
+
+        except (SyntaxError, ValueError) as e:
+            raise ValueError(f"Invalid replacement_pairs format: {e}")
+
+        if axis == "index":
+            df_out = dataframe.rename(index=d)
+        elif axis == "columns":
+            df_out = dataframe.rename(columns=d)
+        else:
+            raise ValueError("Invalid axis. Only 'index' or 'columns' are supported.")
+        
+        return (df_out,)

modules/pandas_wrapper/utils.py

@@ -1,4 +1,4 @@
-from typing import Any, List, Tuple, Union, Set
+from typing import Any, List, Tuple, Union, Set, Optional
 import json
 import numpy as np
 import pandas as pd
@@ -54,7 +54,7 @@ def column_labels_string_to_list(dataframe:pd.DataFrame, s:str, no_check:bool=Fa
     Args:
         dataframe (pd.DataFrame): The DataFrame to validate column names against.
         s (str): A comma-separated string representing column labels.
-        no_check (str): If True, input string is processed as a string without checking the column labels
+        no_check (bool): If True, input string is processed as a string without checking the column labels
                         of the DataFrame.
 
     Returns:
@@ -84,6 +84,48 @@ def column_labels_string_to_list(dataframe:pd.DataFrame, s:str, no_check:bool=Fa
     return output_columns
 
 
+def column_label_string_to_target_type(dataframe: pd.DataFrame, label: str, return_none: bool=False) -> Optional[Union[str, int, float]]:
+    """
+    Converts a string column label into the data type used for the corresponding label in the DataFrame.
+
+    If a column name does not exist in the DataFrame, a ValueError is raised.
+
+    Args:
+        dataframe (pd.DataFrame): The DataFrame to validate column names against.
+        label (str): A string representing a column label.
+        return_none (bool): If True, if the label is not found, None is returned instead of raising an exception.
+    Returns:
+        Union[str, int, float]: The column label in the correct type used in the DataFrame.
+
+    Raises:
+        ValueError: If `s` does not match any existing column name.
+    """
+    valid_columns = flatten_list_tuple_unique(dataframe.columns)
+    s = label.strip()
+
+    if s in valid_columns:
+        return s
+
+    try:
+        numeric_c = int(s)
+        if numeric_c in valid_columns:
+            return numeric_c
+    except ValueError:
+        pass
+
+    try:
+        numeric_c = float(s)
+        if numeric_c in valid_columns:
+            return numeric_c
+    except ValueError:
+        pass
+
+    if return_none:
+        return None
+
+    raise ValueError(f"Column '{s}' not found in the DataFrame.")
+
+
 def series_to_jsons(series: pd.Series) -> str:
     """
     Converts a pandas Series to a JSON string.

modules/seaborn_wrapper/sns_bar.py

@@ -6,7 +6,7 @@ from .common import common_input_types
 from .util import plot_post_steps
 from .util import set_up_sns_style_palette
 from .util import process_y_columns_and_labels
-
+from ..pandas_wrapper.utils import column_label_string_to_target_type
 
 class SNSBar:
     """
@@ -63,6 +63,7 @@ class SNSBar:
         """
         df = dataframe
         set_up_sns_style_palette(style, palette)
+        x_column_name = column_label_string_to_target_type(df, x_column_name)
         y_columns, labels = process_y_columns_and_labels(df, y_column_name, legend_label)
 
         # Create the plot

modules/seaborn_wrapper/sns_boxplot.py

@@ -4,7 +4,7 @@ import pandas as pd
 import seaborn as sns
 from .util import plot_post_steps
 from .util import set_up_sns_style_palette
-
+from ..pandas_wrapper.utils import column_label_string_to_target_type
 
 class SNSBoxplot:
     """
@@ -81,6 +81,8 @@ class SNSBoxplot:
         """
         df = dataframe
         set_up_sns_style_palette(style, palette)
+        category_column_name = column_label_string_to_target_type(df, category_column_name, return_none=True)
+        value_column_name = column_label_string_to_target_type(df, value_column_name)
 
         # Create the plot
         fig, ax = plt.subplots()

modules/seaborn_wrapper/sns_line.py

@@ -6,7 +6,7 @@ from .common import common_input_types
 from .util import plot_post_steps
 from .util import set_up_sns_style_palette
 from .util import process_y_columns_and_labels
-
+from ..pandas_wrapper.utils import column_label_string_to_target_type
 
 class SNSLine:
     """
@@ -83,6 +83,7 @@ class SNSLine:
         """
         df = dataframe
         set_up_sns_style_palette(style, palette)
+        x_column_name = column_label_string_to_target_type(df, x_column_name)
         y_columns, labels = process_y_columns_and_labels(df, y_column_name, legend_label)
 
         fig, ax = plt.subplots()

modules/seaborn_wrapper/sns_pairplot.py

@@ -4,6 +4,7 @@ import pandas as pd
 import seaborn as sns
 from .util import plot_post_steps
 from .util import set_up_sns_style_palette
+from ..pandas_wrapper.utils import column_label_string_to_target_type
 
 
 class SNSPairplot:
@@ -60,6 +61,7 @@ class SNSPairplot:
         set_up_sns_style_palette(style, palette)
 
         # Create the plot
+        hue_column_name = column_label_string_to_target_type(df, hue_column_name, return_none=True)
         if hue_column_name:
             pp = sns.pairplot(df, hue=hue_column_name)
         else:

modules/seaborn_wrapper/sns_scatter.py

@@ -6,7 +6,7 @@ from .common import common_input_types
 from .util import plot_post_steps
 from .util import set_up_sns_style_palette
 from .util import process_y_columns_and_labels
-
+from ..pandas_wrapper.utils import column_label_string_to_target_type
 
 class SNSScatter:
     """
@@ -63,6 +63,7 @@ class SNSScatter:
         """
         df = dataframe
         set_up_sns_style_palette(style, palette)
+        x_column_name = column_label_string_to_target_type(df, x_column_name)
         y_columns, labels = process_y_columns_and_labels(df, y_column_name, legend_label)
 
         # Create the plot