rdt 1.12.1.dev1__tar.gz → 1.12.2.dev0__tar.gz

{rdt-1.12.1.dev1/rdt.egg-info → rdt-1.12.2.dev0}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: rdt
-Version: 1.12.1.dev1
+Version: 1.12.2.dev0
 Summary: Reversible Data Transforms
 Author-email: "DataCebo, Inc." <info@sdv.dev>
 License: BSL-1.1

{rdt-1.12.1.dev1 → rdt-1.12.2.dev0}/pyproject.toml

@@ -137,7 +137,7 @@ collect_ignore = ['pyproject.toml']
 exclude_lines = ['NotImplementedError()']
 
 [tool.bumpversion]
-current_version = "1.12.1.dev1"
+current_version = "1.12.2.dev0"
 parse = '(?P<major>\d+)\.(?P<minor>\d+)\.(?P<patch>\d+)(\.(?P<release>[a-z]+)(?P<candidate>\d+))?'
 serialize = [
     '{major}.{minor}.{patch}.{release}{candidate}',
@@ -178,13 +178,14 @@ preview = true
 line-length = 100
 indent-width = 4
 src = ["rdt"]
-target-version = "py312"
 exclude = [
-    "docs",
-    ".tox",
-    ".git",
-    "__pycache__",
-    ".ipynb_checkpoints"
+    'docs',
+    '.tox',
+    '.git',
+    '__pycache__',
+    '.ipynb_checkpoints',
+    'tasks.py',
+    'tests/contributing.py'
 ]
 
 [tool.ruff.lint]
@@ -194,13 +195,22 @@ select = [
     # Pycodestyle
     "E",
     "W",
+    # pydocstyle
+    "D",
     # isort
-    "I001"
+    "I001",
+    # print statements
+    "T201",
+    # pandas-vet
+    "PD"
 ]
 ignore = [
     "E501",
+    # pydocstyle
     "D107",  # Missing docstring in __init__
     "D417",   # Missing argument descriptions in the docstring, this is a bug from pydocstyle: https://github.com/PyCQA/pydocstyle/issues/449
+    "PD901",
+    "PD101",
 ]
 
 [tool.ruff.format]
@@ -210,14 +220,19 @@ preview = true
 docstring-code-format = true
 docstring-code-line-length = "dynamic"
 
-[tool.ruff.lint.pep8-naming]
-extend-ignore-names = ["X", "C", "X_padded", "Y", "Y_padded"]
-
 [tool.ruff.lint.isort]
 known-first-party = ["rdt"]
+lines-between-types = 0
 
 [tool.ruff.lint.per-file-ignores]
 "__init__.py" = ["F401", "E402", "F403", "F405", "E501", "I001"]
+"errors.py" = ["D105"]
+"tests/**.py" = ["D"]
+"tests/contributing.py" = ["T201"]
 
 [tool.ruff.lint.pydocstyle]
-convention = "google"
+convention = "google"
+
+[tool.ruff.lint.pycodestyle]
+max-doc-length = 100
+max-line-length = 100

{rdt-1.12.1.dev1 → rdt-1.12.2.dev0}/rdt/__init__.py

@@ -4,7 +4,7 @@
 
 __author__ = 'DataCebo, Inc.'
 __email__ = 'info@sdv.dev'
-__version__ = '1.12.1.dev1'
+__version__ = '1.12.2.dev0'
 
 
 import sys

{rdt-1.12.1.dev1 → rdt-1.12.2.dev0}/rdt/transformers/boolean.py

@@ -78,7 +78,7 @@ class BinaryEncoder(BaseTransformer):
             data (pandas.Series):
                 Data to transform.
 
-        Returns
+        Returns:
             np.ndarray
         """
         data = pd.to_numeric(data, errors='coerce')
@@ -110,3 +110,20 @@ class BinaryEncoder(BaseTransformer):
         data[isna] = np.nan
 
         return data
+
+    def _set_fitted_parameters(self, column_name, null_transformer):
+        """Manually set the parameters on the transformer to get it into a fitted state.
+
+        Args:
+            column_name (str):
+                The name of the column to use for the transformer.
+            null_transformer (NullTransformer):
+                A fitted null transformer instance that can be used to generate
+                null values for the column.
+        """
+        self.reset_randomization()
+        self.columns = [column_name]
+        self.output_columns = [column_name]
+        self.null_transformer = null_transformer
+        if self.null_transformer.models_missing_values():
+            self.output_columns.append(column_name + '.is_null')

{rdt-1.12.1.dev1 → rdt-1.12.2.dev0}/rdt/transformers/categorical.py

@@ -135,6 +135,24 @@ class UniformEncoder(BaseTransformer):
 
         self.frequencies, self.intervals = self._compute_frequencies_intervals(labels, freq)
 
+    def _set_fitted_parameters(self, column_name, intervals, dtype='object'):
+        """Manually set the parameters on the transformer to get it into a fitted state.
+
+        Args:
+            column_name (str):
+                The name of the column for this transformer.
+            intervals (dict[str, tuple]):
+                A dictionary mapping categories to the interval in the range [0, 1]
+                it should map to.
+            dtype (str, optional):
+                The dtype to convert the reverse transformed data back to. Defaults to 'object'.
+        """
+        self.reset_randomization()
+        self.columns = [column_name]
+        self.output_columns = [column_name]
+        self.intervals = intervals
+        self.dtype = dtype
+
     def _transform(self, data):
         """Map the category to a continuous value.
 

{rdt-1.12.1.dev1 → rdt-1.12.2.dev0}/rdt/transformers/datetime.py

@@ -5,6 +5,7 @@ import pandas as pd
 from pandas.api.types import is_datetime64_dtype, is_numeric_dtype
 from pandas.core.tools.datetimes import _guess_datetime_format_for_array
 
+from rdt.errors import TransformerInputError
 from rdt.transformers.base import BaseTransformer
 from rdt.transformers.null import NullTransformer
 
@@ -151,6 +152,41 @@ class UnixTimestampEncoder(BaseTransformer):
                 'next_transformer': None,
             }
 
+    def _set_fitted_parameters(
+        self, column_name, null_transformer, min_max_values=None, dtype='object'
+    ):
+        """Manually set the parameters on the transformer to get it into a fitted state.
+
+        Args:
+            column_name (str):
+                The name of the column for this transformer.
+            null_transformer (NullTransformer):
+                A fitted null transformer instance that can be used to generate
+                null values for the column.
+            min_max_values (tuple or None):
+                None or a tuple containing the (min, max) values for the transformer.
+                Should be used to set self._min_value and self._max_value and must be
+                provided if self.enforce_min_max_values is True.
+                Defaults to None.
+            dtype (str, optional):
+                The dtype to convert the reverse transformed data back to. Defaults to 'object'.
+        """
+        self.reset_randomization()
+        self.columns = [column_name]
+        self.output_columns = [column_name]
+        self._dtype = dtype
+
+        if self.enforce_min_max_values and not min_max_values:
+            raise TransformerInputError('Must provide min and max values for this transformer.')
+
+        if min_max_values:
+            self._min_value = min_max_values[0]
+            self._max_value = min_max_values[1]
+
+        self.null_transformer = null_transformer
+        if self.null_transformer.models_missing_values():
+            self.output_columns.append(column_name + '.is_null')
+
     def _transform(self, data):
         """Transform datetime values to float values.
 

{rdt-1.12.1.dev1 → rdt-1.12.2.dev0}/rdt/transformers/null.py

@@ -121,6 +121,20 @@ class NullTransformer:
             if self._missing_value_generation == 'random':
                 self._null_percentage = null_values.sum() / len(data)
 
+    def _set_fitted_parameters(self, null_ratio):
+        """Manually set the parameters on the transformer to get it into a fitted state.
+
+        Args:
+            null_ratio (float):
+                The fraction of values to replace with null values.
+        """
+        if null_ratio < 0 or null_ratio > 1.0:
+            raise ValueError('null_ratio should be a value between 0 and 1.')
+
+        if null_ratio != 0:
+            self.nulls = True
+            self._null_percentage = null_ratio
+
     def transform(self, data):
         """Replace null values with the indicated ``missing_value_replacement``.
 

{rdt-1.12.1.dev1 → rdt-1.12.2.dev0}/rdt/transformers/numerical.py

@@ -8,6 +8,7 @@ import pandas as pd
 import scipy
 from sklearn.mixture import BayesianGaussianMixture
 
+from rdt.errors import TransformerInputError
 from rdt.transformers.base import BaseTransformer
 from rdt.transformers.null import NullTransformer
 from rdt.transformers.utils import learn_rounding_digits
@@ -198,6 +199,49 @@ class FloatFormatter(BaseTransformer):
 
         return data.astype(self._dtype)
 
+    def _set_fitted_parameters(
+        self,
+        column_name,
+        null_transformer,
+        rounding_digits=None,
+        min_max_values=None,
+        dtype='object',
+    ):
+        """Manually set the parameters on the transformer to get it into a fitted state.
+
+        Args:
+            column_name (str):
+                The name of the column to use for the transformer.
+            null_transformer (NullTransformer):
+                A fitted null transformer instance that can be used to generate
+                null values for the column.
+            min_max_values (Tuple(float) or None):
+                None or a tuple containing the (min, max) values for the transformer.
+            rounding_digits (int or None):
+                The number of digits to round to.
+            dtype (str):
+                The pandas dtype the reversed data will be converted into.
+        """
+        self.reset_randomization()
+        self.null_transformer = null_transformer
+        self.columns = [column_name]
+        self.output_columns = [column_name]
+        if self.enforce_min_max_values:
+            if not min_max_values:
+                raise TransformerInputError('Must provide min and max values for this transformer.')
+
+        if min_max_values:
+            self._min_value = min(min_max_values)
+            self._max_value = max(min_max_values)
+
+        if rounding_digits:
+            self._rounding_digits = rounding_digits
+
+        if self.null_transformer.models_missing_values():
+            self.output_columns.append(column_name + '.is_null')
+
+        self._dtype = dtype
+
 
 class GaussianNormalizer(FloatFormatter):
     r"""Transformer for numerical data based on copulas transformation.

{rdt-1.12.1.dev1 → rdt-1.12.2.dev0}/rdt/transformers/pii/anonymizer.py

@@ -308,6 +308,30 @@ class AnonymizedFaker(BaseTransformer):
 
         return reverse_transformed
 
+    def _set_fitted_parameters(self, column_name, nan_frequency=0.0, cardinality=None):
+        """Manually set the parameters on the transformer to get it into a fitted state.
+
+        Args:
+            column_name (str):
+                The name of the column to use for the transformer.
+            nan_frequency (float):
+                The fraction of values that should be replaced with nan values
+                if self.missing_value_generation is 'random'.
+            cardinality (int or None):
+                The number of unique values to generate if cardinality rule is set to
+                'match'.
+        """
+        self.reset_randomization()
+        self.columns = [column_name]
+        self.output_columns = [column_name]
+        if self.cardinality_rule == 'match':
+            if not cardinality:
+                raise TransformerInputError(
+                    'Cardinality "match" rule must specify a cardinality value.'
+                )
+        self._data_cardinality = cardinality
+        self._nan_frequency = nan_frequency
+
     def __repr__(self):
         """Represent initialization of transformer as text.
 

{rdt-1.12.1.dev1 → rdt-1.12.2.dev0/rdt.egg-info}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.1
 Name: rdt
-Version: 1.12.1.dev1
+Version: 1.12.2.dev0
 Summary: Reversible Data Transforms
 Author-email: "DataCebo, Inc." <info@sdv.dev>
 License: BSL-1.1