sdv 1.37.3.dev0__tar.gz → 1.37.4.dev0__tar.gz

{sdv-1.37.3.dev0/sdv.egg-info → sdv-1.37.4.dev0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sdv
-Version: 1.37.3.dev0
+Version: 1.37.4.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
@@ -48,7 +48,7 @@ Requires-Dist: deepecho>=0.7.0; python_version < "3.14"
 Requires-Dist: deepecho>=0.8.0; python_version >= "3.14"
 Requires-Dist: rdt>=1.18.2; python_version < "3.14"
 Requires-Dist: rdt>=1.20.0; python_version >= "3.14"
-Requires-Dist: sdmetrics>=0.28.0
+Requires-Dist: sdmetrics>=0.28.1
 Requires-Dist: platformdirs>=4.0
 Requires-Dist: pyyaml>=6.0.1
 Provides-Extra: excel

{sdv-1.37.3.dev0 → sdv-1.37.4.dev0}/pyproject.toml

@@ -46,7 +46,7 @@ dependencies = [
     "deepecho>=0.8.0;python_version>='3.14'",
     "rdt>=1.18.2;python_version<'3.14'",
     "rdt>=1.20.0;python_version>='3.14'",
-    "sdmetrics>=0.28.0",
+    "sdmetrics>=0.28.1",
     'platformdirs>=4.0',
     'pyyaml>=6.0.1',
 ]
@@ -149,7 +149,7 @@ namespaces = false
 version = {attr = 'sdv.__version__'}
 
 [tool.bumpversion]
-current_version = "1.37.3.dev0"
+current_version = "1.37.4.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-1.37.3.dev0 → sdv-1.37.4.dev0}/sdv/__init__.py

@@ -6,7 +6,7 @@
 
 __author__ = 'DataCebo, Inc.'
 __email__ = 'info@sdv.dev'
-__version__ = '1.37.3.dev0'
+__version__ = '1.37.4.dev0'
 
 
 import sys

{sdv-1.37.3.dev0 → sdv-1.37.4.dev0}/sdv/datasets/demo.py

@@ -640,6 +640,20 @@ def _find_text_key(contents, dataset_prefix, filename):
     return None
 
 
+def _validate_resource_filepath(resource_filepath):
+    """Validate a resource filepath is a string, not undefined and does not begin with slash."""
+    if not isinstance(resource_filepath, str):
+        raise TypeError('`resource_filepath` must be a string.')
+
+    if not resource_filepath:
+        raise ValueError('`resource_filepath` cannot be empty.')
+
+    if resource_filepath.startswith('/'):
+        raise ValueError(
+            "`resource_filepath` must be relative to the dataset and cannot begin with '/'."
+        )
+
+
 def _validate_text_file_content(modality, output_filepath, filename):
     """Validation for the text file content method."""
     _validate_modalities(modality)
@@ -750,7 +764,7 @@ def _save_file_content(
         dataset_name (str):
             The name of the dataset.
         filename (str):
-            The filename to fetch (``'README.txt'`` or ``'SOURCE.txt'``).
+            The filename to fetch (``'README.txt'`` or ``'SOURCE.txt'`, ``'schemas/postgre.sql'``).
         output_filepath (str):
             Save the file contents at this path.
         bucket (str):
@@ -860,10 +874,11 @@ def get_readme(
 def save_resource(
     modality,
     dataset_name,
-    resource_filename,
-    output_filepath,
+    resource_filepath=None,
+    output_filepath=None,
     s3_bucket_name='sdv-datasets-public',
     credentials=None,
+    resource_filename=None,
 ):
     """Save the resource to disk.
 
@@ -871,11 +886,12 @@ def save_resource(
         modality (str):
             The modality of the dataset: ``'single_table'``, ``'multi_table'``, ``'sequential'``.
         dataset_name (str):
-            The name of the dataset to get the README for.
-        resource_filename (str):
-            The name of the file to download from S3.
-        output_filepath (str or None):
-            Optional path where to save the file.
+            The name of the dataset to get the resource for.
+        resource_filepath (str):
+            The location of the file to download from S3. This can be a filename or a filepath.
+            The location of the file should be under the dataset.
+        output_filepath (str):
+            The filepath where the resource should be saved.
         s3_bucket_name (str, optional):
             The name of the bucket to download from. Only 'sdv-datasets-public' is supported in
             SDV Community. SDV Enterprise is required for other buckets. Defaults to
@@ -887,11 +903,36 @@ def save_resource(
                 'license_key': '<MY_LICENSE_KEY>'
             }
             Defaults to None.
+        resource_filename (str, optional):
+            **Deprecated.**
+            The name of the file to download from S3. Use
+            ``resource_filepath`` instead. Defaults to None.
     """
+    if resource_filepath is None and resource_filename is None:
+        raise ValueError('Please provide a `resource_filepath`.')
+
+    if output_filepath is None:
+        raise ValueError('Please provide an `output_filepath`.')
+
+    if resource_filepath and resource_filename:
+        raise ValueError(
+            'Cannot use both `resource_filepath` and `resource_filename`. '
+            'Please use only `resource_filepath`.'
+        )
+
+    if resource_filename is not None:
+        deprecation_msg = (
+            'Warning: The `resource_filename` parameter is deprecated. '
+            'Please use the `resource_filepath` parameter instead.'
+        )
+        warnings.warn(deprecation_msg, FutureWarning)
+        resource_filepath = resource_filename
+
+    _validate_resource_filepath(resource_filepath)
     _save_file_content(
         modality=modality,
         dataset_name=dataset_name,
-        filename=resource_filename,
+        filename=resource_filepath,
         output_filepath=output_filepath,
         bucket=s3_bucket_name,
         credentials=credentials,

sdv-1.37.4.dev0/sdv/evaluation/__init__.py

@@ -0,0 +1,12 @@
+"""Module to compare real and synthetic data."""
+
+from sdv.evaluation.evaluation import (
+    evaluate_quality,
+    run_diagnostic,
+)
+
+
+__all__ = (
+    'evaluate_quality',
+    'run_diagnostic',
+)

sdv-1.37.4.dev0/sdv/evaluation/evaluation.py

@@ -0,0 +1,102 @@
+"""Methods to compare the real and synthetic data."""
+
+import pandas as pd
+from sdmetrics.reports import DiagnosticReport, QualityReport
+
+from sdv.metadata.metadata import Metadata
+
+DEFAULT_SINGLE_TABLE_NAME = 'table'
+ALLOWED_TYPES = (pd.DataFrame, dict)
+
+
+def _validate_data_type(data, argument_name):
+    if not isinstance(data, ALLOWED_TYPES):
+        raise TypeError(
+            f'{argument_name} must be a pandas DataFrame or dictionary, got {type(data).__name__}.'
+        )
+
+
+def _validate_data(real_data, synthetic_data):
+    _validate_data_type(real_data, 'real_data')
+    _validate_data_type(synthetic_data, 'synthetic_data')
+
+    if type(real_data) is not type(synthetic_data):
+        raise TypeError(
+            'real_data and synthetic_data must have the same type. '
+            f'Got {type(real_data).__name__} and '
+            f'{type(synthetic_data).__name__}.'
+        )
+
+
+def _handle_single_table(real_data, synthetic_data, metadata):
+    if isinstance(real_data, pd.DataFrame) and isinstance(synthetic_data, pd.DataFrame):
+        table_name = DEFAULT_SINGLE_TABLE_NAME
+        if isinstance(metadata, Metadata):
+            table_name = metadata._get_single_table_name() or table_name
+        else:
+            metadata = Metadata.load_from_dict(
+                metadata.to_dict(), single_table_name=DEFAULT_SINGLE_TABLE_NAME
+            )
+
+        real_data = {table_name: real_data}
+        synthetic_data = {table_name: synthetic_data}
+
+    return real_data, synthetic_data, metadata
+
+
+def evaluate_quality(real_data, synthetic_data, metadata, verbose=True):
+    """Evaluate the quality of the synthetic data.
+
+    Args:
+        real_data (pd.DataFrame):
+            The table containing the real data.
+        synthetic_data (pd.DataFrame):
+            The table containing the synthetic data.
+        metadata (Metadata):
+            The metadata object describing the real/synthetic data.
+        verbose (bool):
+            Whether or not to print report summary and progress.
+            Defaults to True.
+
+    Returns:
+        QualityReport:
+            Single table quality report object.
+    """
+    _validate_data(real_data, synthetic_data)
+    real_data, synthetic_data, metadata = _handle_single_table(
+        real_data=real_data,
+        synthetic_data=synthetic_data,
+        metadata=metadata,
+    )
+    quality_report = QualityReport()
+    quality_report.generate(real_data, synthetic_data, metadata.to_dict(), verbose)
+    return quality_report
+
+
+def run_diagnostic(real_data, synthetic_data, metadata, verbose=True):
+    """Run diagnostic report for the synthetic data.
+
+    Args:
+        real_data (pd.DataFrame):
+            The table containing the real data.
+        synthetic_data (pd.DataFrame):
+            The table containing the synthetic data.
+        metadata (Metadata):
+            The metadata object describing the real/synthetic data.
+        verbose (bool):
+            Whether or not to print report summary and progress.
+            Defaults to True.
+
+    Returns:
+        DiagnosticReport:
+            Single table diagnostic report object.
+    """
+    _validate_data(real_data, synthetic_data)
+    real_data, synthetic_data, metadata = _handle_single_table(
+        real_data=real_data,
+        synthetic_data=synthetic_data,
+        metadata=metadata,
+    )
+    diagnostic_report = DiagnosticReport()
+    diagnostic_report.generate(real_data, synthetic_data, metadata.to_dict(), verbose)
+    return diagnostic_report

{sdv-1.37.3.dev0 → sdv-1.37.4.dev0}/sdv/evaluation/multi_table.py

@@ -1,5 +1,7 @@
 """Methods to compare the real and synthetic data for multi-table."""
 
+import warnings
+
 from sdmetrics import visualization
 from sdmetrics.reports.multi_table.diagnostic_report import DiagnosticReport
 from sdmetrics.reports.multi_table.quality_report import QualityReport
@@ -7,7 +9,7 @@ from sdmetrics.reports.multi_table.quality_report import QualityReport
 import sdv.evaluation.single_table as single_table_visualization
 
 
-def evaluate_quality(real_data, synthetic_data, metadata, verbose=True):
+def _evaluate_quality(real_data, synthetic_data, metadata, verbose=True):
     """Evaluate the quality of the synthetic data.
 
     Args:
@@ -30,7 +32,7 @@ def evaluate_quality(real_data, synthetic_data, metadata, verbose=True):
     return quality_report
 
 
-def run_diagnostic(real_data, synthetic_data, metadata, verbose=True):
+def _run_diagnostic(real_data, synthetic_data, metadata, verbose=True):
     """Run diagnostic report for the synthetic data.
 
     Args:
@@ -172,3 +174,30 @@ def get_cardinality_plot(
         parent_primary_key,
         plot_type,
     )
+
+
+DEPRECATED_EVALUATION_FUNCTIONS = {
+    'evaluate_quality': _evaluate_quality,
+    'run_diagnostic': _run_diagnostic,
+}
+
+PLOT_FUNCTIONS = {
+    'get_cardinality_plot': get_cardinality_plot,
+    'get_column_pair_plot': get_column_pair_plot,
+    'get_column_plot': get_column_plot,
+}
+
+
+def __getattr__(name):
+    if name in DEPRECATED_EVALUATION_FUNCTIONS:
+        warnings.warn(
+            "The evaluation functions are now accessible via the 'sdv.evaluation' module.",
+            FutureWarning,
+            stacklevel=2,
+        )
+        return DEPRECATED_EVALUATION_FUNCTIONS.get(name)
+
+    if name not in PLOT_FUNCTIONS:
+        raise AttributeError(f'module {__name__!r} has no attribute {name!r}')
+
+    return PLOT_FUNCTIONS.get(name)

{sdv-1.37.3.dev0 → sdv-1.37.4.dev0}/sdv/evaluation/single_table.py

@@ -1,5 +1,7 @@
 """Methods to compare the real and synthetic data for single-table."""
 
+import warnings
+
 from sdmetrics import visualization
 from sdmetrics.reports.single_table.diagnostic_report import DiagnosticReport
 from sdmetrics.reports.single_table.quality_report import QualityReport
@@ -9,7 +11,7 @@ from sdv.evaluation._utils import _prepare_data_visualization
 from sdv.metadata.metadata import Metadata
 
 
-def evaluate_quality(real_data, synthetic_data, metadata, verbose=True):
+def _evaluate_quality(real_data, synthetic_data, metadata, verbose=True):
     """Evaluate the quality of the synthetic data.
 
     Args:
@@ -35,7 +37,7 @@ def evaluate_quality(real_data, synthetic_data, metadata, verbose=True):
     return quality_report
 
 
-def run_diagnostic(real_data, synthetic_data, metadata, verbose=True):
+def _run_diagnostic(real_data, synthetic_data, metadata, verbose=True):
     """Run diagnostic report for the synthetic data.
 
     Args:
@@ -165,3 +167,29 @@ def get_column_pair_plot(
     )
 
     return visualization.get_column_pair_plot(real_data, synthetic_data, column_names, plot_type)
+
+
+DEPRECATED_EVALUATION_FUNCTIONS = {
+    'evaluate_quality': _evaluate_quality,
+    'run_diagnostic': _run_diagnostic,
+}
+
+PLOT_FUNCTIONS = {
+    'get_column_pair_plot': get_column_pair_plot,
+    'get_column_plot': get_column_plot,
+}
+
+
+def __getattr__(name):
+    if name in DEPRECATED_EVALUATION_FUNCTIONS:
+        warnings.warn(
+            "The evaluation functions are now accessible via the 'sdv.evaluation' module.",
+            FutureWarning,
+            stacklevel=2,
+        )
+        return DEPRECATED_EVALUATION_FUNCTIONS.get(name)
+
+    if name not in PLOT_FUNCTIONS:
+        raise AttributeError(f'module {__name__!r} has no attribute {name!r}')
+
+    return PLOT_FUNCTIONS.get(name)

{sdv-1.37.3.dev0 → sdv-1.37.4.dev0/sdv.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: sdv
-Version: 1.37.3.dev0
+Version: 1.37.4.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
@@ -48,7 +48,7 @@ Requires-Dist: deepecho>=0.7.0; python_version < "3.14"
 Requires-Dist: deepecho>=0.8.0; python_version >= "3.14"
 Requires-Dist: rdt>=1.18.2; python_version < "3.14"
 Requires-Dist: rdt>=1.20.0; python_version >= "3.14"
-Requires-Dist: sdmetrics>=0.28.0
+Requires-Dist: sdmetrics>=0.28.1
 Requires-Dist: platformdirs>=4.0
 Requires-Dist: pyyaml>=6.0.1
 Provides-Extra: excel

{sdv-1.37.3.dev0 → sdv-1.37.4.dev0}/sdv.egg-info/SOURCES.txt

@@ -36,6 +36,7 @@ sdv/datasets/demo.py
 sdv/datasets/local.py
 sdv/evaluation/__init__.py
 sdv/evaluation/_utils.py
+sdv/evaluation/evaluation.py
 sdv/evaluation/multi_table.py
 sdv/evaluation/single_table.py
 sdv/io/__init__.py

{sdv-1.37.3.dev0 → sdv-1.37.4.dev0}/sdv.egg-info/requires.txt

@@ -2,7 +2,7 @@ boto3<2.0.0,>=1.28
 botocore<2.0.0,>=1.31
 graphviz>=0.13.2
 tqdm>=4.29
-sdmetrics>=0.28.0
+sdmetrics>=0.28.1
 platformdirs>=4.0
 pyyaml>=6.0.1
 

sdv-1.37.3.dev0/sdv/evaluation/__init__.py

@@ -1 +0,0 @@
-"""Module to compare real and synthetic data."""