feat: add fuction for applying a given transformation

Author: judithspd

.idea/.gitignore

@@ -20,6 +20,7 @@
     apply_hierarchy,
     check_gen_level,
     get_transformation,
+    apply_transformation,
     generate_intervals,
 )
 
@@ -28,5 +29,6 @@
     "apply_hierarchy",
     "check_gen_level",
     "get_transformation",
+    "apply_transformation",
     "generate_intervals",
 ]

.idea/anjana.iml

@@ -79,6 +79,45 @@ def apply_hierarchy(
     return data_anon
 
 
+@beartype()
+def apply_hierarchy_current(
+    data: typing.Union[typing.List, np.ndarray],
+    hierarchies: dict,
+    level: int,
+    actual: int,
+) -> typing.Union[typing.List, np.ndarray]:
+    """Apply certain level of a hierarchy for a quasi-identifier given the current level.
+
+    :param data: data under study.
+    :type data: list, numpy array
+
+    :param hierarchies: hierarchies for generalizing a given QI.
+    :type hierarchies: dictionary with the hierarchies and the levels
+
+    :param level: level of the hierarchy to be applied.
+    :type level: int
+
+    :param actual: current level of the hierarchy applied.
+    :type actual: int
+
+    :return: column with the given level of hierarchy applied.
+    :rtype: numpy array
+    """
+    num_level = len(hierarchies.keys()) - 1
+    if level > num_level:
+        raise ValueError("Error, invalid hierarchy level")
+    if not isinstance(hierarchies[level], pd.Series):
+        hierarchies[level] = pd.Series(hierarchies[level])
+    if not isinstance(hierarchies[actual], pd.Series):
+        hierarchies[actual] = pd.Series(hierarchies[actual])
+
+    pos = []
+    for elem in data:
+        pos.append(np.where(hierarchies[actual].values == elem)[0][0])
+    data_anon = hierarchies[level].values[pos]
+    return data_anon
+
+
 @beartype()
 def check_gen_level(
     data: pd.DataFrame,
@@ -120,7 +159,7 @@ def get_transformation(
 ) -> list:
     """Get the transformation applied for anonymizing the data.
 
-    Example: a transformation (0,1,2,0) means:
+    Example: a transformation [0,1,2,0] means:
     - Level 0 of generalization for th 1st QI
     - Level 1 of generalization for th 2nd QI
     - Level 2 of generalization for th 3rd QI
@@ -151,6 +190,48 @@ def get_transformation(
     return transformation
 
 
+@beartype()
+def apply_transformation(
+    data_anon: pd.DataFrame,
+    quasi_ident: typing.Union[typing.List, np.ndarray],
+    hierarchies: dict,
+    transformation: list,
+) -> pd.DataFrame:
+    """Apply a given transformation to the data.
+
+    :param data_anon: data under study.
+    :type data_anon: pandas dataframe
+
+    :param quasi_ident: list with the name of the columns of the dataframe
+        that are quasi-identifiers.
+    :type quasi_ident: list of strings
+
+    :param hierarchies: hierarchies for generalizing the QI.
+    :type hierarchies: dictionary containing one dictionary for QI
+        with the hierarchies and the levels
+
+    :param transformation: transformation to be applied
+    :type transformation: list
+
+    :return: dataset generalized with the transformation given
+    :rtype: pandas dataframe
+    """
+    actual_transform = check_gen_level(data_anon, quasi_ident, hierarchies)
+    for i, qi in enumerate(quasi_ident):
+        hierarchy_qi = hierarchies[qi]
+        level = transformation[i]
+        if level > max(hierarchies[qi].keys()):
+            raise ValueError("Error, invalid hierarchy level")
+        actual = actual_transform[qi]
+        if level != actual:
+            column = apply_hierarchy_current(
+                data_anon[qi].values, hierarchy_qi, level, actual
+            )
+            data_anon[qi] = column
+
+    return data_anon
+
+
 @beartype()
 def generate_intervals(
     quasi_ident: typing.Union[typing.List, np.ndarray],

.idea/inspectionProfiles/Project_Default.xml

@@ -75,3 +75,18 @@
 print(transformation_raw)  # [0, 0, 0]
 transformation_anon = utils.get_transformation(data_anon, quasi_ident, hierarchies)
 print(transformation_anon)  # [2, 0, 1]
+
+# Testing the function apply_transformation
+data_transform1 = utils.apply_transformation(data, quasi_ident, hierarchies, [1, 1, 1])
+print(data_transform1)
+print(utils.get_transformation(data_transform1, quasi_ident, hierarchies))  # [1, 1, 1]
+
+data_transform2 = utils.apply_transformation(data, quasi_ident, hierarchies, [5, 1, 1])
+print(data_transform2)
+print(utils.get_transformation(data_transform2, quasi_ident, hierarchies))  # [5, 1, 1]
+
+data_transform3 = utils.apply_transformation(
+    data_anon, quasi_ident, hierarchies, [5, 1, 1]
+)
+print(data_transform3)
+print(utils.get_transformation(data_transform3, quasi_ident, hierarchies))  # [5, 1, 1]