rdt 1.10.1.dev0__tar.gz → 1.11.0.dev0__tar.gz

{rdt-1.10.1.dev0/rdt.egg-info → rdt-1.11.0.dev0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: rdt
-Version: 1.10.1.dev0
+Version: 1.11.0.dev0
 Summary: Reversible Data Transforms
 Author-email: "DataCebo, Inc." <info@sdv.dev>
 License: BSL-1.1
@@ -19,23 +19,27 @@ Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Requires-Python: <3.12,>=3.8
+Requires-Python: <3.13,>=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: numpy<2,>=1.20.0; python_version < "3.10"
-Requires-Dist: numpy<2,>=1.23.3; python_version >= "3.10"
+Requires-Dist: numpy>=1.20.0; python_version < "3.10"
+Requires-Dist: numpy>=1.23.3; python_version >= "3.10" and python_version < "3.12"
+Requires-Dist: numpy>=1.26.0; python_version >= "3.12"
 Requires-Dist: pandas>=1.1.3; python_version < "3.10"
 Requires-Dist: pandas>=1.3.4; python_version >= "3.10" and python_version < "3.11"
 Requires-Dist: pandas>=1.5.0; python_version >= "3.11"
-Requires-Dist: scipy<2,>=1.5.4; python_version < "3.10"
-Requires-Dist: scipy<2,>=1.9.2; python_version >= "3.10"
-Requires-Dist: scikit-learn<2,>=0.24; python_version < "3.10"
+Requires-Dist: scipy>=1.5.4; python_version < "3.10"
+Requires-Dist: scipy>=1.9.2; python_version >= "3.10" and python_version < "3.12"
+Requires-Dist: scipy>=1.12.0; python_version >= "3.12"
+Requires-Dist: scikit-learn>=0.24; python_version < "3.10"
 Requires-Dist: scikit-learn>=1.1.0; python_version >= "3.10" and python_version < "3.11"
-Requires-Dist: scikit-learn<2,>=1.1.3; python_version >= "3.11"
-Requires-Dist: Faker<20,>=17
+Requires-Dist: scikit-learn>=1.1.3; python_version >= "3.11" and python_version < "3.12"
+Requires-Dist: scikit-learn>=1.3.0; python_version >= "3.12"
+Requires-Dist: Faker>=17
 Provides-Extra: copulas
-Requires-Dist: copulas<0.11,>=0.10.0; extra == "copulas"
+Requires-Dist: copulas>=0.11.0.dev0; extra == "copulas"
 Provides-Extra: test
 Requires-Dist: rdt[copulas]; extra == "test"
 Requires-Dist: pytest>=3.4.2; extra == "test"
@@ -50,7 +54,7 @@ Requires-Dist: rdt[test]; extra == "dev"
 Requires-Dist: build<2,>=1.0.0; extra == "dev"
 Requires-Dist: bump-my-version<1,>=0.18.3; extra == "dev"
 Requires-Dist: pip>=9.0.1; extra == "dev"
-Requires-Dist: watchdog<0.11,>=0.8.3; extra == "dev"
+Requires-Dist: watchdog<5,>=1.0.1; extra == "dev"
 Requires-Dist: pycodestyle<2.8.0,>=2.7.0; extra == "dev"
 Requires-Dist: pyflakes<2.4.0,>=2.3.0; extra == "dev"
 Requires-Dist: flake8<4,>=3.7.7; extra == "dev"

{rdt-1.10.1.dev0 → rdt-1.11.0.dev0}/pyproject.toml

@@ -12,25 +12,29 @@ classifiers = [
     'Programming Language :: Python :: 3.9',
     'Programming Language :: Python :: 3.10',
     'Programming Language :: Python :: 3.11',
+    'Programming Language :: Python :: 3.12',
     'Topic :: Scientific/Engineering :: Artificial Intelligence',
 ]
 keywords = ['machine learning', 'synthetic data generation', 'benchmark', 'generative models']
 dynamic = ['version']
 license = { text = 'BSL-1.1' }
-requires-python = '>=3.8,<3.12'
+requires-python = '>=3.8,<3.13'
 readme = 'README.md'
 dependencies = [
-    "numpy>=1.20.0,<2;python_version<'3.10'",
-    "numpy>=1.23.3,<2;python_version>='3.10'",
+    "numpy>=1.20.0;python_version<'3.10'",
+    "numpy>=1.23.3;python_version>='3.10' and python_version<'3.12'",
+    "numpy>=1.26.0;python_version>='3.12'",
     "pandas>=1.1.3;python_version<'3.10'",
     "pandas>=1.3.4;python_version>='3.10' and python_version<'3.11'",
     "pandas>=1.5.0;python_version>='3.11'",
-    "scipy>=1.5.4,<2;python_version<'3.10'",
-    "scipy>=1.9.2,<2;python_version>='3.10'",
-    "scikit-learn>=0.24,<2;python_version<'3.10'",
+    "scipy>=1.5.4;python_version<'3.10'",
+    "scipy>=1.9.2;python_version>='3.10' and python_version<'3.12'",
+    "scipy>=1.12.0;python_version>='3.12'",
+    "scikit-learn>=0.24;python_version<'3.10'",
     "scikit-learn>=1.1.0;python_version>='3.10' and python_version<'3.11'",
-    "scikit-learn>=1.1.3,<2;python_version>='3.11'",
-    'Faker>=17,<20',
+    "scikit-learn>=1.1.3;python_version>='3.11' and python_version<'3.12'",
+    "scikit-learn>=1.3.0;python_version>='3.12'",
+    'Faker>=17',
 ]
 
 [project.urls]
@@ -44,7 +48,7 @@ dependencies = [
 rdt = { main = 'rdt.cli.__main__:main' }
 
 [project.optional-dependencies]
-copulas = ['copulas>=0.10.0,<0.11',]
+copulas = ['copulas>=0.11.0.dev0',]
 test = [
     'rdt[copulas]',
 
@@ -63,7 +67,7 @@ dev = [
     'build>=1.0.0,<2',
     'bump-my-version>=0.18.3,<1',
     'pip>=9.0.1',
-    'watchdog>=0.8.3,<0.11',
+    'watchdog>=1.0.1,<5',
 
     # style check
     'pycodestyle<2.8.0,>=2.7.0',
@@ -133,12 +137,12 @@ namespaces = false
     '*.png',
     '*.gif'
 ]
-'tests' = ['*']
 
 [tool.setuptools.exclude-package-data]
 '*' = [
     '* __pycache__',
     '*.py[co]',
+    'static_code_analysis.txt',
 ]
 
 [tool.setuptools.dynamic]
@@ -162,7 +166,7 @@ collect_ignore = ['pyproject.toml']
 exclude_lines = ['NotImplementedError()']
 
 [tool.bumpversion]
-current_version = "1.10.1.dev0"
+current_version = "1.11.0.dev0"
 parse = '(?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(\.(?P<release>[a-z]+)(?P<candidate>\d+))?'
 serialize = [
     '{major}.{minor}.{patch}.{release}{candidate}',

{rdt-1.10.1.dev0 → rdt-1.11.0.dev0}/rdt/__init__.py

@@ -5,17 +5,17 @@
 
 __author__ = 'DataCebo, Inc.'
 __email__ = 'info@sdv.dev'
-__version__ = '1.10.1.dev0'
+__version__ = '1.11.0.dev0'
 
 
 import sys
 import warnings
+from importlib.metadata import entry_points
 from operator import attrgetter
 from types import ModuleType
 
 import numpy as np
 import pandas as pd
-from pkg_resources import iter_entry_points
 
 from rdt import transformers
 from rdt.hyper_transformer import HyperTransformer
@@ -149,7 +149,13 @@ def _find_addons():
     from top_module.addon_module import x
     """
     group = 'rdt_modules'
-    for entry_point in iter_entry_points(group=group):
+    try:
+        eps = entry_points(group=group)  # pylint: disable=E1123
+    except TypeError:
+        # Load-time selection requires Python >= 3.10 or importlib_metadata >= 3.6
+        eps = entry_points().get(group, [])
+
+    for entry_point in eps:
         try:
             addon = entry_point.load()
         except Exception:  # pylint: disable=broad-exception-caught

rdt-1.11.0.dev0/rdt/transformers/_validators.py

@@ -0,0 +1,171 @@
+"""Validations for multi-column transformers."""
+import importlib
+
+from rdt.errors import TransformerInputError
+
+
+class BaseValidator:
+    """Base validation class.
+
+    The validation classes ensure that the input data is compatible with the transformers
+    and that they can be imported.
+    """
+
+    SUPPORTED_SDTYPES = []
+    VALIDATION_TYPE = None
+
+    @classmethod
+    def _validate_supported_sdtypes(cls, columns_to_sdtypes):
+        message = ''
+        for column, sdtype in columns_to_sdtypes.items():
+            if sdtype not in cls.SUPPORTED_SDTYPES:
+                message += f"Column '{column}' has an unsupported sdtype '{sdtype}'.\n"
+
+        if message:
+            message += (
+                f'Please provide a column that is compatible with {cls.VALIDATION_TYPE} data.'
+            )
+            raise TransformerInputError(message)
+
+    @classmethod
+    def validate_sdtypes(cls, columns_to_sdtypes):
+        """Validate the columns to sdtypes mapping.
+
+        This method aims to call all other sdtype validation method in the class.
+
+        Args:
+            columns_to_sdtypes (dict):
+                Mapping of column names to sdtypes.
+        """
+        raise NotImplementedError
+
+    @classmethod
+    def validate_imports(cls):
+        """Check that the transformers can be imported."""
+        raise NotImplementedError
+
+    @classmethod
+    def validate(cls, columns_to_sdtypes):
+        """Validate the input data.
+
+        Args:
+            columns_to_sdtypes (dict):
+                Mapping of column names to sdtypes.
+        """
+        cls.validate_sdtypes(columns_to_sdtypes)
+        cls.validate_imports()
+
+
+class AddressValidator(BaseValidator):
+    """Validation class for Address data."""
+
+    SUPPORTED_SDTYPES = [
+        'country_code', 'administrative_unit', 'city', 'postcode',
+        'street_address', 'secondary_address', 'state', 'state_abbr'
+    ]
+    VALIDATION_TYPE = 'Address'
+
+    @classmethod
+    def _validate_number_columns(cls, columns_to_sdtypes):
+        if len(columns_to_sdtypes) > 7:
+            raise TransformerInputError(
+                f'{cls.VALIDATION_TYPE} transformers takes up to 7 columns to transform. '
+                'Please provide address data with valid fields.'
+            )
+
+    @staticmethod
+    def _validate_uniqueness_sdtype(columns_to_sdtypes):
+        sdtypes_to_columns = {}
+        for column, sdtype in columns_to_sdtypes.items():
+            if sdtype not in sdtypes_to_columns:
+                sdtypes_to_columns[sdtype] = []
+
+            sdtypes_to_columns[sdtype].append(column)
+
+        duplicate_fields = {
+            value: keys for value, keys in sdtypes_to_columns.items() if len(keys) > 1
+        }
+
+        if duplicate_fields:
+            message = ''
+            for sdtype, columns in duplicate_fields.items():
+                to_print = "', '".join(columns)
+                message += f"Columns '{to_print}' have the same sdtype '{sdtype}'.\n"
+
+            message += 'Your address data cannot have duplicate fields.'
+            raise TransformerInputError(message)
+
+    @classmethod
+    def _validate_administrative_unit(cls, columns_to_sdtypes):
+        num_column_administrative_unit = sum(
+            1 for itm in columns_to_sdtypes.values() if itm in ['administrative_unit', 'state']
+        )
+        if num_column_administrative_unit > 1:
+            raise TransformerInputError(
+                f"The {cls.__name__} can have up to 1 column with sdtype 'state'"
+                f" or 'administrative_unit'. Please provide address data with valid fields."
+            )
+
+    @classmethod
+    def validate_sdtypes(cls, columns_to_sdtypes):
+        """Validate the columns to sdtypes mapping."""
+        cls._validate_supported_sdtypes(columns_to_sdtypes)
+        cls._validate_number_columns(columns_to_sdtypes)
+        cls._validate_uniqueness_sdtype(columns_to_sdtypes)
+        cls._validate_administrative_unit(columns_to_sdtypes)
+
+    @classmethod
+    def validate_imports(cls):
+        """Check that the address transformers can be imported."""
+        error_message = (
+            'You must have SDV Enterprise with the address add-on to use the address features.'
+        )
+
+        try:
+            address_module = importlib.import_module('rdt.transformers.address')
+        except ModuleNotFoundError:
+            raise ImportError(error_message) from None
+
+        required_classes = ['RandomLocationGenerator', 'RegionalAnonymizer']
+        for class_name in required_classes:
+            if not hasattr(address_module, class_name):
+                raise ImportError(error_message)
+
+
+class GPSValidator(BaseValidator):
+    """Validation class for GPS data."""
+
+    SUPPORTED_SDTYPES = ['latitude', 'longitude']
+    VALIDATION_TYPE = 'GPS'
+
+    @staticmethod
+    def _validate_uniqueness_sdtype(columns_to_sdtypes):
+        sdtypes_to_columns = {sdtype: column for column, sdtype in columns_to_sdtypes.items()}
+        if len(sdtypes_to_columns) != 2:
+            raise TransformerInputError(
+                'The GPS columns must have one latitude and on longitude columns sdtypes. '
+                'Please provide GPS data with valid fields.'
+            )
+
+    @classmethod
+    def validate_sdtypes(cls, columns_to_sdtypes):
+        """Validate the columns to sdtypes mapping."""
+        cls._validate_supported_sdtypes(columns_to_sdtypes)
+        cls._validate_uniqueness_sdtype(columns_to_sdtypes)
+
+    @classmethod
+    def validate_imports(cls):
+        """Check that the GPS transformers can be imported."""
+        error_message = (
+            'You must have SDV Enterprise with the gps add-on to use the GPS features.'
+        )
+
+        try:
+            gps_module = importlib.import_module('rdt.transformers.gps')
+        except ModuleNotFoundError:
+            raise ImportError(error_message) from None
+
+        required_classes = ['RandomLocationGenerator', 'MetroAreaAnonymizer', 'GPSNoiser']
+        for class_name in required_classes:
+            if not hasattr(gps_module, class_name):
+                raise ImportError(error_message)

{rdt-1.10.1.dev0 → rdt-1.11.0.dev0}/rdt/transformers/pii/anonymizer.py

@@ -138,7 +138,7 @@ class AnonymizedFaker(BaseTransformer):
         self._faker_random_seed = None
         self.locales = locales
         self.faker = faker.Faker(self.locales)
-        if self.locales:
+        if self.provider_name != 'BaseProvider' and self.locales:
             self._check_locales()
 
         if missing_value_generation not in ['random', None]:

{rdt-1.10.1.dev0 → rdt-1.11.0.dev0/rdt.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: rdt
-Version: 1.10.1.dev0
+Version: 1.11.0.dev0
 Summary: Reversible Data Transforms
 Author-email: "DataCebo, Inc." <info@sdv.dev>
 License: BSL-1.1
@@ -19,23 +19,27 @@ Classifier: Programming Language :: Python :: 3.8
 Classifier: Programming Language :: Python :: 3.9
 Classifier: Programming Language :: Python :: 3.10
 Classifier: Programming Language :: Python :: 3.11
+Classifier: Programming Language :: Python :: 3.12
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Requires-Python: <3.12,>=3.8
+Requires-Python: <3.13,>=3.8
 Description-Content-Type: text/markdown
 License-File: LICENSE
-Requires-Dist: numpy<2,>=1.20.0; python_version < "3.10"
-Requires-Dist: numpy<2,>=1.23.3; python_version >= "3.10"
+Requires-Dist: numpy>=1.20.0; python_version < "3.10"
+Requires-Dist: numpy>=1.23.3; python_version >= "3.10" and python_version < "3.12"
+Requires-Dist: numpy>=1.26.0; python_version >= "3.12"
 Requires-Dist: pandas>=1.1.3; python_version < "3.10"
 Requires-Dist: pandas>=1.3.4; python_version >= "3.10" and python_version < "3.11"
 Requires-Dist: pandas>=1.5.0; python_version >= "3.11"
-Requires-Dist: scipy<2,>=1.5.4; python_version < "3.10"
-Requires-Dist: scipy<2,>=1.9.2; python_version >= "3.10"
-Requires-Dist: scikit-learn<2,>=0.24; python_version < "3.10"
+Requires-Dist: scipy>=1.5.4; python_version < "3.10"
+Requires-Dist: scipy>=1.9.2; python_version >= "3.10" and python_version < "3.12"
+Requires-Dist: scipy>=1.12.0; python_version >= "3.12"
+Requires-Dist: scikit-learn>=0.24; python_version < "3.10"
 Requires-Dist: scikit-learn>=1.1.0; python_version >= "3.10" and python_version < "3.11"
-Requires-Dist: scikit-learn<2,>=1.1.3; python_version >= "3.11"
-Requires-Dist: Faker<20,>=17
+Requires-Dist: scikit-learn>=1.1.3; python_version >= "3.11" and python_version < "3.12"
+Requires-Dist: scikit-learn>=1.3.0; python_version >= "3.12"
+Requires-Dist: Faker>=17
 Provides-Extra: copulas
-Requires-Dist: copulas<0.11,>=0.10.0; extra == "copulas"
+Requires-Dist: copulas>=0.11.0.dev0; extra == "copulas"
 Provides-Extra: test
 Requires-Dist: rdt[copulas]; extra == "test"
 Requires-Dist: pytest>=3.4.2; extra == "test"
@@ -50,7 +54,7 @@ Requires-Dist: rdt[test]; extra == "dev"
 Requires-Dist: build<2,>=1.0.0; extra == "dev"
 Requires-Dist: bump-my-version<1,>=0.18.3; extra == "dev"
 Requires-Dist: pip>=9.0.1; extra == "dev"
-Requires-Dist: watchdog<0.11,>=0.8.3; extra == "dev"
+Requires-Dist: watchdog<5,>=1.0.1; extra == "dev"
 Requires-Dist: pycodestyle<2.8.0,>=2.7.0; extra == "dev"
 Requires-Dist: pyflakes<2.4.0,>=2.3.0; extra == "dev"
 Requires-Dist: flake8<4,>=3.7.7; extra == "dev"

{rdt-1.10.1.dev0 → rdt-1.11.0.dev0}/rdt.egg-info/SOURCES.txt

@@ -24,6 +24,7 @@ rdt/performance/datasets/pii.py
 rdt/performance/datasets/text.py
 rdt/performance/datasets/utils.py
 rdt/transformers/__init__.py
+rdt/transformers/_validators.py
 rdt/transformers/base.py
 rdt/transformers/boolean.py
 rdt/transformers/categorical.py

{rdt-1.10.1.dev0 → rdt-1.11.0.dev0}/rdt.egg-info/requires.txt

@@ -1,32 +1,39 @@
-Faker<20,>=17
+Faker>=17
 
 [:python_version < "3.10"]
-numpy<2,>=1.20.0
+numpy>=1.20.0
 pandas>=1.1.3
-scipy<2,>=1.5.4
-scikit-learn<2,>=0.24
-
-[:python_version >= "3.10"]
-numpy<2,>=1.23.3
-scipy<2,>=1.9.2
+scipy>=1.5.4
+scikit-learn>=0.24
 
 [:python_version >= "3.10" and python_version < "3.11"]
 pandas>=1.3.4
 scikit-learn>=1.1.0
 
+[:python_version >= "3.10" and python_version < "3.12"]
+numpy>=1.23.3
+scipy>=1.9.2
+
 [:python_version >= "3.11"]
 pandas>=1.5.0
-scikit-learn<2,>=1.1.3
+
+[:python_version >= "3.11" and python_version < "3.12"]
+scikit-learn>=1.1.3
+
+[:python_version >= "3.12"]
+numpy>=1.26.0
+scipy>=1.12.0
+scikit-learn>=1.3.0
 
 [copulas]
-copulas<0.11,>=0.10.0
+copulas>=0.11.0.dev0
 
 [dev]
 rdt[test]
 build<2,>=1.0.0
 bump-my-version<1,>=0.18.3
 pip>=9.0.1
-watchdog<0.11,>=0.8.3
+watchdog<5,>=1.0.1
 pycodestyle<2.8.0,>=2.7.0
 pyflakes<2.4.0,>=2.3.0
 flake8<4,>=3.7.7

{rdt-1.10.1.dev0 → rdt-1.11.0.dev0}/tests/test_tasks.py

@@ -1,4 +1,5 @@
 """Tests for the ``tasks.py`` file."""
+
 from tasks import _get_minimum_versions
 
 
@@ -15,7 +16,7 @@ def test_get_minimum_versions():
         "pandas>=1.2.0,<2;python_version<'3.10'",
         "pandas>=1.3.0,<2;python_version>='3.10'",
         'humanfriendly>=8.2,<11',
-        'pandas @ git+https://github.com/pandas-dev/pandas.git@master#egg=pandas'
+        'pandas @ git+https://github.com/pandas-dev/pandas.git@master',
     ]
 
     # Run
@@ -25,12 +26,12 @@ def test_get_minimum_versions():
     # Assert
     expected_versions_39 = [
         'numpy==1.20.0',
-        'pandas @ git+https://github.com/pandas-dev/pandas.git@master#egg=pandas',
+        'git+https://github.com/pandas-dev/pandas.git@master#egg=pandas',
         'humanfriendly==8.2',
     ]
     expected_versions_310 = [
         'numpy==1.23.3',
-        'pandas @ git+https://github.com/pandas-dev/pandas.git@master#egg=pandas',
+        'git+https://github.com/pandas-dev/pandas.git@master#egg=pandas',
         'humanfriendly==8.2',
     ]