sdv 2.0.0.dev0__tar.gz → 2.0.1.dev0__tar.gz

{sdv-2.0.0.dev0/sdv.egg-info → sdv-2.0.1.dev0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sdv
-Version: 2.0.0.dev0
+Version: 2.0.1.dev0
 Summary: Generate synthetic data for single table, multi table and sequential data
 Author-email: "DataCebo, Inc." <info@sdv.dev>
 License-Expression: BUSL-1.1
@@ -14,14 +14,13 @@ Classifier: Development Status :: 5 - Production/Stable
 Classifier: Intended Audience :: Developers
 Classifier: Natural Language :: English
 Classifier: Programming Language :: Python :: 3
-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: Programming Language :: Python :: 3.13
 Classifier: Programming Language :: Python :: 3.14
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Requires-Python: <3.15,>=3.9
+Requires-Python: <3.15,>=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: boto3<2.0.0,>=1.28
@@ -29,8 +28,7 @@ Requires-Dist: botocore<2.0.0,>=1.31
 Requires-Dist: cloudpickle>=2.1.0; python_version < "3.14"
 Requires-Dist: cloudpickle>=3.1.1; python_version >= "3.14"
 Requires-Dist: graphviz>=0.13.2
-Requires-Dist: numpy>=1.22.2; python_version < "3.10"
-Requires-Dist: numpy>=1.24.0; python_version >= "3.10" and python_version < "3.12"
+Requires-Dist: numpy>=1.24.0; python_version < "3.12"
 Requires-Dist: numpy>=1.26.0; python_version >= "3.12" and python_version < "3.13"
 Requires-Dist: numpy>=2.1.0; python_version >= "3.13" and python_version < "3.14"
 Requires-Dist: numpy>=2.3.2; python_version >= "3.14"
@@ -46,7 +44,7 @@ Requires-Dist: ctgan>=0.11.1; python_version < "3.14"
 Requires-Dist: ctgan>=0.12.0; python_version >= "3.14"
 Requires-Dist: deepecho>=0.7.0; python_version < "3.14"
 Requires-Dist: deepecho>=0.8.0; python_version >= "3.14"
-Requires-Dist: rdt>=2.0.0.dev0
+Requires-Dist: rdt>=2.0.0
 Requires-Dist: sdmetrics>=0.32.0
 Requires-Dist: platformdirs>=4.0
 Requires-Dist: pyyaml>=6.0.1
@@ -64,7 +62,6 @@ Requires-Dist: google-api-python-client; extra == "test"
 Requires-Dist: google-auth; extra == "test"
 Requires-Dist: google-auth-oauthlib; extra == "test"
 Requires-Dist: requests; extra == "test"
-Requires-Dist: pyarrow; extra == "test"
 Requires-Dist: gitpython; extra == "test"
 Requires-Dist: slack-sdk<4.0,>=3.23; extra == "test"
 Provides-Extra: pomegranate
@@ -91,6 +88,8 @@ Requires-Dist: coverage<8,>=4.5.12; extra == "dev"
 Requires-Dist: invoke; extra == "dev"
 Provides-Extra: readme
 Requires-Dist: rundoc<0.5,>=0.4.3; extra == "readme"
+Provides-Extra: benchmark
+Requires-Dist: pyarrow; extra == "benchmark"
 Dynamic: license-file
 
 <div align="center">

{sdv-2.0.0.dev0 → sdv-2.0.1.dev0}/pyproject.toml

@@ -7,7 +7,6 @@ classifiers = [
     'Intended Audience :: Developers',
     'Natural Language :: English',
     'Programming Language :: Python :: 3',
-    'Programming Language :: Python :: 3.9',
     'Programming Language :: Python :: 3.10',
     'Programming Language :: Python :: 3.11',
     'Programming Language :: Python :: 3.12',
@@ -19,7 +18,7 @@ keywords = ['sdv', 'synthetic-data', 'synthetic-data-generation', 'timeseries',
 dynamic = ['version']
 license = 'BUSL-1.1'
 license-files = ['LICENSE']
-requires-python = '>=3.9,<3.15'
+requires-python = '>=3.10,<3.15'
 readme = 'README.md'
 dependencies = [
     'boto3>=1.28,<2.0.0',
@@ -27,8 +26,7 @@ dependencies = [
     "cloudpickle>=2.1.0;python_version<'3.14'",
     "cloudpickle>=3.1.1;python_version>='3.14'",
     'graphviz>=0.13.2',
-    "numpy>=1.22.2;python_version<'3.10'",
-    "numpy>=1.24.0;python_version>='3.10' and python_version<'3.12'",
+    "numpy>=1.24.0;python_version<'3.12'",
     "numpy>=1.26.0;python_version>='3.12' and python_version<'3.13'",
     "numpy>=2.1.0;python_version>='3.13' and python_version<'3.14'",
     "numpy>=2.3.2;python_version>='3.14'",
@@ -44,7 +42,7 @@ dependencies = [
     "ctgan>=0.12.0;python_version>='3.14'",
     "deepecho>=0.7.0;python_version<'3.14'",
     "deepecho>=0.8.0;python_version>='3.14'",
-    "rdt>=2.0.0.dev0",
+    "rdt>=2.0.0",
     "sdmetrics>=0.32.0",
     'platformdirs>=4.0',
     'pyyaml>=6.0.1',
@@ -74,7 +72,6 @@ test = [
     'google-auth',
     'google-auth-oauthlib',
     'requests',
-    'pyarrow',
     'gitpython',
     'slack-sdk>=3.23,<4.0',
 ]
@@ -113,6 +110,7 @@ dev = [
     'invoke'
 ]
 readme = ['rundoc>=0.4.3,<0.5',]
+benchmark = ['pyarrow']
 
 [tool.setuptools]
 include-package-data = true
@@ -148,7 +146,7 @@ namespaces = false
 version = {attr = 'sdv.__version__'}
 
 [tool.bumpversion]
-current_version = "2.0.0.dev0"
+current_version = "2.0.1.dev0"
 parse = '(?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(\.(?P<release>[a-z]+)(?P<candidate>\d+))?'
 serialize = [
     '{major}.{minor}.{patch}.{release}{candidate}',

{sdv-2.0.0.dev0 → sdv-2.0.1.dev0}/sdv/__init__.py

@@ -6,7 +6,7 @@
 
 __author__ = 'DataCebo, Inc.'
 __email__ = 'info@sdv.dev'
-__version__ = '2.0.0.dev0'
+__version__ = '2.0.1.dev0'
 
 
 import sys

{sdv-2.0.0.dev0 → sdv-2.0.1.dev0}/sdv/cag/base.py

@@ -90,8 +90,15 @@ class BaseConstraint:
                 f'{missing_attrs} are not saved as attributes on the constraint.'
             )
 
+        if self._is_single_table and getattr(self, '_table_name', None) is not None:
+            instanced['table_name'] = self._table_name
+
         return {'class_name': self.__class__.__name__, 'parameters': instanced}
 
+    def to_dict(self):
+        """Return the constraint as a serializable dictionary."""
+        return self.get_constraint_dict()
+
     @classmethod
     def load_constraint_from_dict(cls, parameters):
         """Uses the given parameters to recreate an instance of the constraint."""

{sdv-2.0.0.dev0 → sdv-2.0.1.dev0}/sdv/cag/programmable_constraint.py

@@ -41,6 +41,10 @@ class ProgrammableConstraint:
             'parameters': instanced,
         }
 
+    def to_dict(self):
+        """Return the constraint as a serializable dictionary."""
+        return self.get_constraint_dict()
+
     @classmethod
     def load_constraint_from_dict(cls, parameters):
         """Uses the given parameters to recreate an instance of the constraint."""

{sdv-2.0.0.dev0 → sdv-2.0.1.dev0}/sdv/multi_table/base.py

@@ -574,7 +574,7 @@ class BaseMultiTableSynthesizer:
                 If ``table_name`` is not present in the metadata.
         """
         self._validate_table_name(table_name)
-        return self._table_synthesizers[table_name].get_transformers()
+        return self._table_synthesizers[table_name].get_transformers(table_name=table_name)
 
     def update_transformers(self, table_name, column_name_to_transformer):
         """Update any of the transformers assigned to each of the column names.
@@ -590,7 +590,9 @@ class BaseMultiTableSynthesizer:
                 If ``table_name`` is not present in the metadata.
         """
         self._validate_table_name(table_name)
-        self._table_synthesizers[table_name].update_transformers(column_name_to_transformer)
+        self._table_synthesizers[table_name].update_transformers(
+            column_name_to_transformer, table_name=table_name
+        )
 
     def _store_and_convert_original_cols(self, data):
         list_of_changed_tables = []
@@ -756,8 +758,8 @@ class BaseMultiTableSynthesizer:
 
     def _validate_sample_input(
         self,
-        table_name,
         num_rows,
+        main_table_name,
         batch_size,
         max_tries_per_batch,
         output_folder_path,
@@ -769,9 +771,17 @@ class BaseMultiTableSynthesizer:
                 'sampling synthetic data.'
             )
 
+        if main_table_name is None:
+            raise SynthesizerInputError(
+                '`main_table_name` cannot be None for multi-table sampling. '
+                'Please provide a table name.'
+            )
+
         table_names = list(self.get_metadata().tables)
-        if table_name not in table_names:
-            raise SynthesizerInputError(f"Table '{table_name}' does not exist in the metadata.")
+        if main_table_name not in table_names:
+            raise SynthesizerInputError(
+                f"Table '{main_table_name}' does not exist in the metadata."
+            )
 
         _validate_positive_integer('num_rows', num_rows)
         _validate_positive_integer('max_tries_per_batch', max_tries_per_batch)
@@ -818,35 +828,7 @@ class BaseMultiTableSynthesizer:
 
         return pd.concat(sampled, ignore_index=True)
 
-    def sample(
-        self,
-        table_name,
-        num_rows,
-        batch_size=None,
-        max_tries_per_batch=100,
-        output_folder_path=None,
-    ):
-        """Generate synthetic data for the entire dataset.
-
-        Args:
-            table_name (str):
-                The name of the main table to sample.
-            num_rows (int):
-                The number of rows to sample.
-            batch_size (int, optional):
-                The batch size for sampling. Defaults to None.
-            max_tries_per_batch (int, optional):
-                The maximum number of tries per batch. Defaults to 100.
-            output_folder_path (str, optional):
-                The folder path to save the sampled data. Defaults to None.
-
-        Returns:
-            dict: A dictionary containing the sampled data for each table.
-        """
-        self._validate_sample_input(
-            table_name, num_rows, batch_size, max_tries_per_batch, output_folder_path
-        )
-        scale = self._resolve_scale(table_name, num_rows)
+    def _sample_from_scale(self, scale, batch_size, max_tries_per_batch, output_folder_path):
         with self._set_temp_numpy_seed(), disable_single_table_logger():
             sampled_data = self._sample(
                 scale=scale,
@@ -878,6 +860,102 @@ class BaseMultiTableSynthesizer:
         if output_folder_path is not None:
             self._save_sampled_data(sampled_data, output_folder_path)
 
+        return sampled_data, total_rows, total_columns
+
+    def sample_scale(
+        self,
+        scale=1.0,
+        batch_size=None,
+        max_tries_per_batch=100,
+        output_folder_path=None,
+    ):
+        """Generate synthetic data for the entire dataset.
+
+        Args:
+            scale (float):
+                A float representing how much to scale the data by. If scale is set to ``1.0``,
+                this does not scale the sizes of the tables. If ``scale`` is greater than ``1.0``
+                create more rows than the original data by a factor of ``scale``.
+                If ``scale`` is lower than ``1.0`` create fewer rows by the factor of ``scale``
+                than the original tables. Defaults to ``1.0``.
+            batch_size (int, optional):
+                The batch size for sampling. Defaults to None.
+            max_tries_per_batch (int, optional):
+                The maximum number of tries per batch. Defaults to 100.
+            output_folder_path (str, optional):
+                The folder path to save the sampled data. Defaults to None.
+
+        Returns:
+            dict: A dictionary containing the sampled data for each table.
+        """
+        if not self._fitted:
+            raise SamplingError(
+                'This synthesizer has not been fitted. Please fit your synthesizer first before '
+                'sampling synthetic data.'
+            )
+
+        if type(scale) not in (float, int) or not scale > 0:
+            raise SynthesizerInputError(
+                f"Invalid parameter for 'scale' ({scale}). Please provide a number that is >0.0."
+            )
+        self._validate_sample_input(
+            num_rows=1,
+            main_table_name=list(self.metadata.tables)[0],
+            batch_size=batch_size,
+            max_tries_per_batch=max_tries_per_batch,
+            output_folder_path=output_folder_path,
+        )
+
+        sampled_data, total_rows, total_columns = self._sample_from_scale(
+            scale, batch_size, max_tries_per_batch, output_folder_path
+        )
+
+        SYNTHESIZER_LOGGER.info({
+            'EVENT': 'Sample',
+            'TIMESTAMP': datetime.datetime.now(),
+            'SYNTHESIZER CLASS NAME': self.__class__.__name__,
+            'SYNTHESIZER ID': self._synthesizer_id,
+            'TOTAL NUMBER OF TABLES': len(sampled_data),
+            'TOTAL NUMBER OF ROWS': total_rows,
+            'TOTAL NUMBER OF COLUMNS': total_columns,
+        })
+
+        return sampled_data
+
+    def sample(
+        self,
+        num_rows,
+        main_table_name=None,
+        batch_size=None,
+        max_tries_per_batch=100,
+        output_folder_path=None,
+    ):
+        """Generate synthetic data for the entire dataset.
+
+        Args:
+            num_rows (int):
+                The number of rows to sample.
+            main_table_name (str):
+                The name of the main table to sample.
+            batch_size (int, optional):
+                The batch size for sampling. Defaults to None.
+            max_tries_per_batch (int, optional):
+                The maximum number of tries per batch. Defaults to 100.
+            output_folder_path (str, optional):
+                The folder path to save the sampled data. Defaults to None.
+
+        Returns:
+            dict: A dictionary containing the sampled data for each table.
+        """
+        self._validate_sample_input(
+            num_rows, main_table_name, batch_size, max_tries_per_batch, output_folder_path
+        )
+        scale = self._resolve_scale(main_table_name, num_rows)
+
+        sampled_data, total_rows, total_columns = self._sample_from_scale(
+            scale, batch_size, max_tries_per_batch, output_folder_path
+        )
+
         SYNTHESIZER_LOGGER.info({
             'EVENT': 'Sample',
             'TIMESTAMP': datetime.datetime.now(),

{sdv-2.0.0.dev0 → sdv-2.0.1.dev0}/sdv/sequential/par.py

@@ -427,17 +427,25 @@ class PARSynthesizer(LossValuesMixin, MissingModuleMixin, BaseSynthesizer):
 
         return processed_table_data
 
-    def update_transformers(self, column_name_to_transformer):
+    def update_transformers(self, column_name_to_transformer, table_name=None):
         """Update any of the transformers assigned to each of the column names.
 
         Args:
             column_name_to_transformer (dict):
                 Dict mapping column names to transformers to be used for that column.
+            table_name (str, optional):
+                The name of the table for which to update the transformers. If not provided,
+                will use the default table name associated with the synthesizer.
 
         Raises:
             ValueError:
                 Raise when the transformer of a context column is passed.
         """
+        if table_name is not None:
+            self._validate_table_name(table_name)
+        else:
+            table_name = self._table_name
+
         forbidden_updates = []
         for column in set(column_name_to_transformer).intersection(set(self.context_columns)):
             if self._get_table_metadata().columns[column]['sdtype'] in MODELABLE_SDTYPES:
@@ -448,7 +456,7 @@ class PARSynthesizer(LossValuesMixin, MissingModuleMixin, BaseSynthesizer):
                 'Transformers for context columns are not allowed to be updated.'
             )
 
-        super().update_transformers(column_name_to_transformer)
+        super().update_transformers(column_name_to_transformer, table_name=table_name)
 
     def _fit_context_model(self, processed_table_data):
         LOGGER.debug(f'Fitting context synthesizer {self._context_synthesizer.__class__.__name__}')
@@ -639,8 +647,8 @@ class PARSynthesizer(LossValuesMixin, MissingModuleMixin, BaseSynthesizer):
 
     def sample(
         self,
-        table_name,
         num_rows,
+        main_table_name=None,
         max_tries_per_batch=100,
         batch_size=None,
         output_folder_path=None,
@@ -651,10 +659,10 @@ class PARSynthesizer(LossValuesMixin, MissingModuleMixin, BaseSynthesizer):
         ``sample_sequences`` instead.
 
         Args:
-            table_name (str):
-                Name of the table to sample.
             num_rows (int):
                 Number of rows to sample.
+            main_table_name (str or None):
+                Name of the table to sample.
             max_tries_per_batch (int):
                 Number of times to retry sampling discarded rows. Defaults to 100.
             batch_size (int or None):
@@ -725,8 +733,8 @@ class PARSynthesizer(LossValuesMixin, MissingModuleMixin, BaseSynthesizer):
         condition_columns = context_columns[condition_columns].to_dict('records')
         synthesizer_conditions = [Condition(conditions) for conditions in condition_columns]
         context = self._context_synthesizer.sample_from_conditions(synthesizer_conditions)
-        context.update(context_columns)
-        return self._sample(context, sequence_length)
+        context[self._table_name].update(context_columns)
+        return {self._table_name: self._sample(context[self._table_name], sequence_length)}
 
     def _process_context_columns(self, context_columns):
         """Process context columns by applying appropriate transformations.

{sdv-2.0.0.dev0 → sdv-2.0.1.dev0}/sdv/single_table/base.py

@@ -212,6 +212,13 @@ class BaseSynthesizer:
             'SYNTHESIZER ID': self._synthesizer_id,
         })
 
+    def _validate_table_name(self, table_name):
+        if table_name != self._table_name:
+            raise ValueError(
+                'The provided table name does not match the metadata:'
+                f"\nTable '{table_name}' is not present in the metadata."
+            )
+
     def _validate_metadata(self, table_data):
         """Validate that the table data follows the metadata."""
         errors = []
@@ -293,13 +300,21 @@ class BaseSynthesizer:
                     'parameter to False.'
                 )
 
-    def update_transformers(self, column_name_to_transformer):
+    def update_transformers(self, column_name_to_transformer, table_name=None):
         """Update any of the transformers assigned to each of the column names.
 
         Args:
             column_name_to_transformer (dict):
                 Dict mapping column names to transformers to be used for that column.
+            table_name (str, optional):
+                The name of the table for which to update the transformers. If not provided,
+                will use the default table name associated with the synthesizer. Defaults to None.
         """
+        if table_name is not None:
+            self._validate_table_name(table_name)
+        else:
+            table_name = self._table_name
+
         self._validate_transformers(column_name_to_transformer)
         self._warn_quality_and_performance(column_name_to_transformer)
         self._warn_unable_to_enforce_rounding(column_name_to_transformer)
@@ -358,16 +373,26 @@ class BaseSynthesizer:
         self.validate(data)
         self._auto_assign_transformers(data)
 
-    def get_transformers(self):
+    def get_transformers(self, table_name=None):
         """Get a dictionary mapping of ``column_name``  and ``rdt.transformers``.
 
         A dictionary representing the column names and the transformers that will be used
         to transform the data.
 
+        Args:
+            table_name (str, optional):
+                The name of the table for which to get the transformers. If not provided,
+                the default table name will be used. Defaults to None.
+
         Returns:
             dict:
                 A dictionary mapping with column names and transformers.
         """
+        if table_name is not None:
+            self._validate_table_name(table_name)
+        else:
+            table_name = self._table_name
+
         field_transformers = self._data_processor._hyper_transformer.field_transformers
         if field_transformers == {}:
             raise ValueError(
@@ -1179,17 +1204,10 @@ class BaseSingleTableSynthesizer(BaseSynthesizer):
                 ' sampling synthetic data.'
             )
 
-    def _validate_table_name(self, table_name):
-        if table_name != self._table_name:
-            raise ValueError(
-                'The provided table name does not match the metadata:'
-                f"\nTable '{table_name}' is not present in the metadata."
-            )
-
     def sample(
         self,
-        table_name,
         num_rows,
+        main_table_name=None,
         max_tries_per_batch=100,
         batch_size=None,
         output_folder_path=None,
@@ -1197,10 +1215,10 @@ class BaseSingleTableSynthesizer(BaseSynthesizer):
         """Sample rows from this table.
 
         Args:
-            table_name (str):
-                Name of the table to sample. This parameter is required.
             num_rows (int):
                 Number of rows to sample. This parameter is required.
+            main_table_name (str):
+                Name of the table to sample.
             max_tries_per_batch (int):
                 Number of times to retry sampling until the batch size is met. Defaults to 100.
             batch_size (int or None):
@@ -1213,6 +1231,7 @@ class BaseSingleTableSynthesizer(BaseSynthesizer):
             dict[str, pandas.DataFrame]:
                 Dictionary mapping the table name to the sampled data.
         """
+        table_name = self._table_name if main_table_name is None else main_table_name
         self._validate_table_name(table_name)
         self._validate_fit_before_sample()
         self._check_input_metadata_updated()
@@ -1422,8 +1441,9 @@ class BaseSingleTableSynthesizer(BaseSynthesizer):
                 The folder to periodically write sampled rows to. Defaults to None.
 
         Returns:
-            pandas.DataFrame:
-                Sampled data.
+            dict[str, pandas.DataFrame]:
+                Dictionary mapping the table name to the sampled data.
+
 
         Raises:
             ConstraintsNotMetError:
@@ -1483,7 +1503,7 @@ class BaseSingleTableSynthesizer(BaseSynthesizer):
             'TOTAL NUMBER OF COLUMNS': len(sampled.columns),
         })
 
-        return sampled
+        return {self._table_name: sampled}
 
     def _validate_known_columns(self, conditions):
         """Validate the user-passed conditions."""
@@ -1507,8 +1527,8 @@ class BaseSingleTableSynthesizer(BaseSynthesizer):
                 The folder to periodically write sampled rows to. Defaults to None.
 
         Returns:
-            pandas.DataFrame:
-                Sampled data.
+            dict[str, pandas.DataFrame]:
+                Dictionary mapping the table name to the sampled data.
 
         Raises:
             ConstraintsNotMetError:
@@ -1557,4 +1577,4 @@ class BaseSingleTableSynthesizer(BaseSynthesizer):
             'TOTAL NUMBER OF COLUMNS': len(sampled.columns),
         })
 
-        return sampled
+        return {self._table_name: sampled}

{sdv-2.0.0.dev0 → sdv-2.0.1.dev0/sdv.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sdv
-Version: 2.0.0.dev0
+Version: 2.0.1.dev0
 Summary: Generate synthetic data for single table, multi table and sequential data
 Author-email: "DataCebo, Inc." <info@sdv.dev>
 License-Expression: BUSL-1.1
@@ -14,14 +14,13 @@ Classifier: Development Status :: 5 - Production/Stable
 Classifier: Intended Audience :: Developers
 Classifier: Natural Language :: English
 Classifier: Programming Language :: Python :: 3
-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: Programming Language :: Python :: 3.13
 Classifier: Programming Language :: Python :: 3.14
 Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
-Requires-Python: <3.15,>=3.9
+Requires-Python: <3.15,>=3.10
 Description-Content-Type: text/markdown
 License-File: LICENSE
 Requires-Dist: boto3<2.0.0,>=1.28
@@ -29,8 +28,7 @@ Requires-Dist: botocore<2.0.0,>=1.31
 Requires-Dist: cloudpickle>=2.1.0; python_version < "3.14"
 Requires-Dist: cloudpickle>=3.1.1; python_version >= "3.14"
 Requires-Dist: graphviz>=0.13.2
-Requires-Dist: numpy>=1.22.2; python_version < "3.10"
-Requires-Dist: numpy>=1.24.0; python_version >= "3.10" and python_version < "3.12"
+Requires-Dist: numpy>=1.24.0; python_version < "3.12"
 Requires-Dist: numpy>=1.26.0; python_version >= "3.12" and python_version < "3.13"
 Requires-Dist: numpy>=2.1.0; python_version >= "3.13" and python_version < "3.14"
 Requires-Dist: numpy>=2.3.2; python_version >= "3.14"
@@ -46,7 +44,7 @@ Requires-Dist: ctgan>=0.11.1; python_version < "3.14"
 Requires-Dist: ctgan>=0.12.0; python_version >= "3.14"
 Requires-Dist: deepecho>=0.7.0; python_version < "3.14"
 Requires-Dist: deepecho>=0.8.0; python_version >= "3.14"
-Requires-Dist: rdt>=2.0.0.dev0
+Requires-Dist: rdt>=2.0.0
 Requires-Dist: sdmetrics>=0.32.0
 Requires-Dist: platformdirs>=4.0
 Requires-Dist: pyyaml>=6.0.1
@@ -64,7 +62,6 @@ Requires-Dist: google-api-python-client; extra == "test"
 Requires-Dist: google-auth; extra == "test"
 Requires-Dist: google-auth-oauthlib; extra == "test"
 Requires-Dist: requests; extra == "test"
-Requires-Dist: pyarrow; extra == "test"
 Requires-Dist: gitpython; extra == "test"
 Requires-Dist: slack-sdk<4.0,>=3.23; extra == "test"
 Provides-Extra: pomegranate
@@ -91,6 +88,8 @@ Requires-Dist: coverage<8,>=4.5.12; extra == "dev"
 Requires-Dist: invoke; extra == "dev"
 Provides-Extra: readme
 Requires-Dist: rundoc<0.5,>=0.4.3; extra == "readme"
+Provides-Extra: benchmark
+Requires-Dist: pyarrow; extra == "benchmark"
 Dynamic: license-file
 
 <div align="center">

{sdv-2.0.0.dev0 → sdv-2.0.1.dev0}/sdv.egg-info/requires.txt

@@ -2,26 +2,23 @@ boto3<2.0.0,>=1.28
 botocore<2.0.0,>=1.31
 graphviz>=0.13.2
 tqdm>=4.29
-rdt>=2.0.0.dev0
+rdt>=2.0.0
 sdmetrics>=0.32.0
 platformdirs>=4.0
 pyyaml>=6.0.1
 
-[:python_version < "3.10"]
-numpy>=1.22.2
-
 [:python_version < "3.11"]
 pandas<3,>=1.4.0
 
+[:python_version < "3.12"]
+numpy>=1.24.0
+
 [:python_version < "3.14"]
 cloudpickle>=2.1.0
 copulas>=0.12.1
 ctgan>=0.11.1
 deepecho>=0.7.0
 
-[:python_version >= "3.10" and python_version < "3.12"]
-numpy>=1.24.0
-
 [:python_version >= "3.11" and python_version < "3.12"]
 pandas<3,>=1.5.0
 
@@ -41,6 +38,9 @@ copulas>=0.14.0
 ctgan>=0.12.0
 deepecho>=0.8.0
 
+[benchmark]
+pyarrow
+
 [dev]
 sdv[test]
 build<2,>=1.0.0
@@ -83,6 +83,5 @@ google-api-python-client
 google-auth
 google-auth-oauthlib
 requests
-pyarrow
 gitpython
 slack-sdk<4.0,>=3.23