feat: __repr__ method to Schema, Row, Column, SupervisedDataset & Tab…

…le; summary() to Column & Table (#333)

Closes #319.

### Summary of Changes

- Added `__str__()`, `__repr__()` and `_ipython_display_()` methods to
the classes `Table`, `Column`, `Row`, `TableSchema`and
`SupervisedDataset`.
- Added the method `summary()` to the `Table` class
- Changed the raised error in some of the ColumnStatistics methods from
`TypeError` to `NonNumericColumnError`
- Moved the methods `idness()` and `stability()` from the class `Column`
to `ColumnStatistics`
  - Refactored the test accordingly

### Testing Instructions

Create an instance of one of the mentioned classes and either call the
print method on it or use the jupyter notebook interactive display
method to visualize it in a prettier way.

Co-authored-by: Marvin Walter <walter@zbmed.de>
Co-authored-by: GideonKoenig <GideonKoenig@users.noreply.github.com>
Co-authored-by: WinPlay02 <winplay02_gh@woberlaender.de>
Co-authored-by: SmiteDeluxe <smitedeluxe@gmail.com>
Co-authored-by: SmiteDeluxe <SmiteDeluxe@users.noreply.github.com>

Runtime/safe-ds/pyproject.toml

@@ -10,6 +10,7 @@ python = "^3.10"
 pandas = "^1.5.3"
 scikit-learn = "^1.2.0"
 seaborn = "^0.12.2"
+ipython = "^8.8.0"
 matplotlib = "^3.6.3"
 
 [tool.poetry.dev-dependencies]

Runtime/safe-ds/safe_ds/data/_column.py

@@ -6,6 +6,7 @@
 
 import numpy as np
 import pandas as pd
+from IPython.core.display_functions import DisplayHandle, display
 from safe_ds.exceptions import (
     ColumnLengthMismatchError,
     ColumnSizeError,
@@ -73,24 +74,6 @@ def get_value(self, index: int) -> Any:
 
         return self._data[index]
 
-    def idness(self) -> float:
-        """
-        Calculates the idness of this column (number of unique values / number of rows).
-
-        Returns
-        -------
-        idness: float
-            The idness of the column
-
-        Raises
-        ------
-        ColumnSizeError
-            If this column is empty
-        """
-        if self._data.size == 0:
-            raise ColumnSizeError("> 0", "0")
-        return self._data.nunique() / self._data.size
-
     @property
     def statistics(self) -> ColumnStatistics:
         return ColumnStatistics(self)
@@ -204,25 +187,6 @@ def has_missing_values(self) -> bool:
             or (isinstance(value, Number) and np.isnan(value))
         )
 
-    def stability(self) -> float:
-        """
-        Calculates the stability of this column.
-        The value is calculated as the ratio between the number of mode values and the number of non-null-values.
-
-        Returns
-        -------
-        stability: float
-            Stability of this column
-
-        Raises
-        ------
-        ColumnSizeError
-            If this column is empty
-        """
-        if self._data.size == 0:
-            raise ColumnSizeError("> 0", "0")
-        return self._data.value_counts()[self.statistics.mode()] / self._data.count()
-
     def correlation_with(self, other_column: Column) -> float:
         """
         Calculates Pearson correlation between this and another column, if both are numerical
@@ -270,6 +234,33 @@ def __eq__(self, other: object) -> bool:
     def __hash__(self) -> int:
         return hash(self._data)
 
+    def __str__(self) -> str:
+        tmp = self._data.to_frame()
+        tmp.columns = [self.name]
+        return tmp.__str__()
+
+    def __repr__(self) -> str:
+        tmp = self._data.to_frame()
+        tmp.columns = [self.name]
+        return tmp.__repr__()
+
+    def _ipython_display_(self) -> DisplayHandle:
+        """
+        Returns a pretty display object for the Column to be used in Jupyter Notebooks
+
+        Returns
+        -------
+        output: DisplayHandle
+            Output object
+        """
+        tmp = self._data.to_frame()
+        tmp.columns = [self.name]
+
+        with pd.option_context(
+            "display.max_rows", tmp.shape[0], "display.max_columns", tmp.shape[1]
+        ):
+            return display(tmp)
+
 
 class ColumnStatistics:
     def __init__(self, column: Column):
@@ -286,11 +277,13 @@ def max(self) -> float:
 
         Raises
         ------
-        TypeError
+        NonNumericColumnError
             If the data contains non-numerical data.
         """
         if not self.column._type.is_numeric():
-            raise TypeError("The column contains non numerical data.")
+            raise NonNumericColumnError(
+                f"{self.column.name} is of type {self.column._type}."
+            )
         return self.column._data.max()
 
     def min(self) -> float:
@@ -304,11 +297,13 @@ def min(self) -> float:
 
         Raises
         ------
-        TypeError
+        NonNumericColumnError
             If the data contains non-numerical data.
         """
         if not self.column._type.is_numeric():
-            raise TypeError("The column contains non numerical data.")
+            raise NonNumericColumnError(
+                f"{self.column.name} is of type {self.column._type}."
+            )
         return self.column._data.min()
 
     def mean(self) -> float:
@@ -322,11 +317,13 @@ def mean(self) -> float:
 
         Raises
         ------
-        TypeError
+        NonNumericColumnError
             If the data contains non-numerical data.
         """
         if not self.column._type.is_numeric():
-            raise TypeError("The column contains non numerical data.")
+            raise NonNumericColumnError(
+                f"{self.column.name} is of type {self.column._type}."
+            )
         return self.column._data.mean()
 
     def mode(self) -> Any:
@@ -351,11 +348,13 @@ def median(self) -> float:
 
         Raises
         ------
-        TypeError
+        NonNumericColumnError
             If the data contains non-numerical data.
         """
         if not self.column._type.is_numeric():
-            raise TypeError("The column contains non numerical data.")
+            raise NonNumericColumnError(
+                f"{self.column.name} is of type {self.column._type}."
+            )
         return self.column._data.median()
 
     def sum(self) -> float:
@@ -369,7 +368,7 @@ def sum(self) -> float:
 
         Raises
         ---
-        NonNumericalColumnError
+        NonNumericColumnError
             If the data is non numerical
 
         """
@@ -391,7 +390,7 @@ def variance(self) -> float:
 
         Raises
         ---
-        NonNumericalColumnError
+        NonNumericColumnError
             If the data is non numerical
 
         """
@@ -414,7 +413,7 @@ def standard_deviation(self) -> float:
 
         Raises
         ---
-        NonNumericalColumnError
+        NonNumericColumnError
             If the data is non numerical
 
         """
@@ -423,3 +422,43 @@ def standard_deviation(self) -> float:
                 f"{self.column.name} is of type {self.column._type}."
             )
         return self.column._data.std()
+
+    def stability(self) -> float:
+        """
+        Calculates the stability of this column.
+        The value is calculated as the ratio between the number of mode values and the number of non-null-values.
+
+        Returns
+        -------
+        stability: float
+            Stability of this column
+
+        Raises
+        ------
+        ColumnSizeError
+            If this column is empty
+        """
+        if self.column._data.size == 0:
+            raise ColumnSizeError("> 0", "0")
+        return (
+            self.column._data.value_counts()[self.column.statistics.mode()]
+            / self.column._data.count()
+        )
+
+    def idness(self) -> float:
+        """
+        Calculates the idness of this column (number of unique values / number of rows).
+
+        Returns
+        -------
+        idness: float
+            The idness of the column
+
+        Raises
+        ------
+        ColumnSizeError
+            If this column is empty
+        """
+        if self.column._data.size == 0:
+            raise ColumnSizeError("> 0", "0")
+        return self.column._data.nunique() / self.column._data.size

Runtime/safe-ds/safe_ds/data/_row.py

@@ -2,8 +2,10 @@
 from typing import Any
 
 import pandas as pd
+from IPython.core.display_functions import DisplayHandle, display
 from safe_ds.exceptions import UnknownColumnNameError
 
+from ._column_type import ColumnType
 from ._table_schema import TableSchema
 
 
@@ -35,7 +37,8 @@ def get_value(self, column_name: str) -> Any:
 
     def has_column(self, column_name: str) -> bool:
         """
-        Returns if the row contains a given column
+        Alias for self.schema.hasColumn(column_name: str) -> bool.
+        Returns if the row contains a given column.
 
         Parameters
         ----------
@@ -49,6 +52,40 @@ def has_column(self, column_name: str) -> bool:
         """
         return self.schema.has_column(column_name)
 
+    def get_column_names(self) -> list[str]:
+        """
+        Alias for self.schema.get_column_names() -> list[str].
+        Returns a list of all column names saved in this schema
+
+        Returns
+        -------
+        column_names: list[str]
+            the column names
+        """
+        return self.schema.get_column_names()
+
+    def get_type_of_column(self, column_name: str) -> ColumnType:
+        """
+        Alias for self.schema.get_type_of_column(column_name: str) -> ColumnType.
+        Returns the type of the given column.
+
+        Parameters
+        ----------
+        column_name : str
+            The name of the column you want the type of
+
+        Returns
+        -------
+        type: ColumnType
+            The type of the column
+
+        Raises
+        ------
+        ColumnNameError
+            If the specified target column name doesn't exist
+        """
+        return self.schema.get_type_of_column(column_name)
+
     def __eq__(self, other: typing.Any) -> bool:
         if not isinstance(other, Row):
             return NotImplemented
@@ -58,3 +95,30 @@ def __eq__(self, other: typing.Any) -> bool:
 
     def __hash__(self) -> int:
         return hash(self._data)
+
+    def __str__(self) -> str:
+        tmp = self._data.to_frame().T
+        tmp.columns = self.get_column_names()
+        return tmp.__str__()
+
+    def __repr__(self) -> str:
+        tmp = self._data.to_frame().T
+        tmp.columns = self.get_column_names()
+        return tmp.__repr__()
+
+    def _ipython_display_(self) -> DisplayHandle:
+        """
+        Returns a pretty display object for the Row to be used in Jupyter Notebooks
+
+        Returns
+        -------
+        output: DisplayHandle
+            Output object
+        """
+        tmp = self._data.to_frame().T
+        tmp.columns = self.get_column_names()
+
+        with pd.option_context(
+            "display.max_rows", tmp.shape[0], "display.max_columns", tmp.shape[1]
+        ):
+            return display(tmp)

Runtime/safe-ds/safe_ds/data/_supervised_dataset.py

@@ -1,3 +1,5 @@
+from IPython.core.display_functions import DisplayHandle
+
 from ._column import Column
 from ._table import Table
 
@@ -26,3 +28,27 @@ def feature_vectors(self) -> Table:
     @property
     def target_values(self) -> Column:
         return self._y
+
+    def __repr__(self) -> str:
+        tmp = self._X.add_column(self._y)
+        header_info = "Target Column is '" + self._y.name + "'\n"
+        return header_info + tmp.__repr__()
+
+    def __str__(self) -> str:
+        tmp = self._X.add_column(self._y)
+        header_info = "Target Column is '" + self._y.name + "'\n"
+        return header_info + tmp.__str__()
+
+    def _ipython_display_(self) -> DisplayHandle:
+        """
+        Returns a pretty display object for the Table to be used in Jupyter Notebooks
+
+        Returns
+        -------
+        output: DisplayHandle
+            Output object
+        """
+        tmp = self._X.add_column(self._y)
+        header_info = "Target Column is '" + self._y.name + "'\n"
+        print(header_info)
+        return tmp._ipython_display_()