Added Pandas Group By node.

__init__.py

@@ -6,6 +6,7 @@ from .modules.matplotlib.mpl_scatter import MPLScatter
 from .modules.pandas_wrapper.pandas_columns import PandasColumns
 from .modules.pandas_wrapper.pandas_create import PandasCreate
 from .modules.pandas_wrapper.pandas_create_with_index import PandasCreateWithIndex
+from .modules.pandas_wrapper.pandas_group_by import PandasGroupBy
 from .modules.pandas_wrapper.pandas_head import PandasHead
 from .modules.pandas_wrapper.pandas_iloc_row_series import PandasIlocRowSeries
 from .modules.pandas_wrapper.pandas_iloc_rows_dataframe import PandasIlocRowsDataFrame
@@ -36,6 +37,7 @@ NODE_CLASS_MAPPINGS: Dict[str, Type[T]] = {
     "PandasColumns": PandasColumns,
     "PandasCreate": PandasCreate,
     "PandasCreateWithIndex": PandasCreateWithIndex,
+    "PandasGroupBy": PandasGroupBy,
     "PandasHead": PandasHead,
     "PandasIlocRowSeries": PandasIlocRowSeries,
     "PandasIlocRowsDataFrame": PandasIlocRowsDataFrame,
@@ -69,6 +71,7 @@ NODE_DISPLAY_NAME_MAPPINGS: Dict[str, str] = {
     "PandasColumns": "Pandas Columns",
     "PandasCreate": "Pandas Create",
     "PandasCreateWithIndex": "Pandas Create With Index",
+    "PandasGroupBy": "Pandas Group By",
     "PandasHead": "Pandas Head",
     "PandasIlocRowSeries": "Pandas Iloc Row Series",
     "PandasIlocRowsDataFrame": "Pandas Iloc Rows DataFrame",

docs/ug.md

@@ -36,9 +36,10 @@ You can right-click, select **Add Node**, go to **Data Analysis**, and look for
 | **Pandas Loc Cell Str**  | Select a cell by row label (index) and column label |
 | **Pandas Head**         | Extract the first few rows         |
 
-**Manipulating data**
+**Transforming data**
 | Node Name               | Functionality                        |
 |-------------------------|-------------------------------------|
+| **Pandas Group By**     | Group rows and apply aggregation functions |
 | **Pandas Join**         | Join two DataFrames                |
 | **Pandas Sort**         | Sort DataFrame by a column         |
 
@@ -120,6 +121,26 @@ Use **Pandas Loc Cell Str**  for selecting a cell by row label (index) and colum
 
 This node outputs the cell value as a string irrespective of the actual cell data type.
 
+## Grouping Rows
+Use **Pandas Group By** to group rows and apply aggregation functions.
+Currently below aggregation functions are supported:
+* sum
+* mean
+* count
+* std
+* min
+* max
+
+Below shows an example of using sum, mean, count functions:
+![Group By](images/group_by1.png)
+
+Below shows an example of using std, min, max functions:
+
+![Group By](images/group_by2.png)
+
+You can also have multiple columns to apply aggregation functions. In the below example, Qty and Price columns are aggregated for the Fruit column:
+![Group By](images/group_by3.png)
+
 ## Joining DataFrames
 Use **Pandas Join**.  You can select from inner, left, right, outer joins.
 First, use two **Pandas Load CSV** nodes to load two CSVs for the left DataFrame and right DataFrame.

modules/pandas_wrapper/pandas_group_by.py

@@ -0,0 +1,65 @@
+from typing import Any, Dict
+from io import StringIO
+import pandas as pd
+
+
+class PandasGroupBy:
+    """
+    PandasGroupBy:
+        A class for computing aggregate values for groups in a DataFrame.
+    """
+    
+    @classmethod
+    def INPUT_TYPES(cls) -> Dict[str, Any]:
+        """
+        Defines the input types for the function.
+
+        See https://pandas.pydata.org/docs/reference/groupby.html for aggragate functions.
+        Returns:
+            Dict[str, Any]: A dictionary specifying required input types.
+        """
+        return {
+            "required": {
+                "dataframe_json": ("STRING", {"multiline": True}),
+                "column_name": ("STRING", {"default": ""}),
+                "aggragate_function":(("sum", "mean", "count", "std", "min", "max"),)
+            }
+        }
+
+    RETURN_TYPES: tuple = ("STRING",)
+    FUNCTION: str = "groupby_dataframes"
+    CATEGORY: str = "Data Analysis"
+
+    def groupby_dataframes(self,
+                           dataframe_json: str,
+                           column_name: str,
+                           aggragate_function: str) -> tuple:
+        """
+        Compute aggregate values for groups in a DataFrame
+
+        Args:
+            dataframe_json (str): A JSON string representation of the DataFrame.
+            column_name (str): The name of the column for use as groups.
+            aggregate_function (str): The type of aggregate function to perform.
+
+        Returns:
+            tuple: A tuple containing a JSON string of the merged DataFrame.
+        """
+        # Deserialize JSON string to DataFrame
+        df = pd.read_json(StringIO(dataframe_json))
+
+        if aggragate_function == "max":
+            df_out = df.groupby(column_name).max()
+        elif aggragate_function == "min":
+            df_out = df.groupby(column_name).min()
+        elif aggragate_function == "std":
+            df_out = df.groupby(column_name).std()
+        elif aggragate_function == "count":
+            df_out = df.groupby(column_name).count()
+        elif aggragate_function == "sum":
+            df_out = df.groupby(column_name).sum()
+        elif aggragate_function == "mean":
+            df_out = df.groupby(column_name).mean()
+        else:
+            raise ValueError("Unsupported aggregate function.")
+        return (df_out.to_json(),)

pyproject.toml

@@ -1,7 +1,7 @@
 [project]
 name = "data-analysis"
 description = "Data analysis custom modules for ComfyUI - Use Pandas & Matplotlib within ComfyUI"
-version = "1.0.15"
+version = "1.0.16"
 license = {file = "LICENSE"}
 
 [project.urls]