Add PyDocStyle Support (#951)

Author: juanitorduz

docs/source/conf.py

@@ -1,4 +1,5 @@
 #!/usr/bin/env python3
+"""Sphinx configuration for PyMC Marketing Docs."""
 
 import os
 

docs/source/notebooks/mmm/mmm_time_varying_media_example.ipynb

@@ -4550,8 +4550,7 @@
     "def plot_posterior(\n",
     "    posterior, figsize=(15, 8), path_color=\"blue\", hist_color=\"blue\", **kwargs\n",
     "):\n",
-    "    \"\"\"\n",
-    "    Plot the posterior distribution of a stochastic process.\n",
+    "    \"\"\"Plot the posterior distribution of a stochastic process.\n",
     "\n",
     "    Parameters\n",
     "    ----------\n",
@@ -4565,8 +4564,8 @@
     "        Color of the histogram.\n",
     "    **kwargs\n",
     "        Additional keyword arguments to pass to the plotting functions.\n",
-    "    \"\"\"\n",
     "\n",
+    "    \"\"\"\n",
     "    # Calculate the expected value (mean) across all draws and chains for each date\n",
     "    expected_value = posterior.mean(dim=(\"draw\", \"chain\"))\n",
     "\n",

docs/source/uml/classes_clv.png

@@ -11,6 +11,8 @@
 #   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 #   See the License for the specific language governing permissions and
 #   limitations under the License.
+"""PyMC Marketing."""
+
 from pymc_marketing import clv, mmm
 from pymc_marketing.version import __version__
 

docs/source/uml/classes_mmm.png

@@ -11,6 +11,8 @@
 #   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 #   See the License for the specific language governing permissions and
 #   limitations under the License.
+"""CLV models and utilities."""
+
 from pymc_marketing.clv.models import (
     BetaGeoModel,
     GammaGammaModel,

pymc_marketing/__init__.py

@@ -11,6 +11,8 @@
 #   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 #   See the License for the specific language governing permissions and
 #   limitations under the License.
+"""Distributions for the CLV module."""
+
 import numpy as np
 import pymc as pm
 import pytensor.tensor as pt
@@ -80,9 +82,9 @@ def _supp_shape_from_params(*args, **kwargs):
 
 
 class ContNonContract(PositiveContinuous):
-    r"""
-    Individual-level model for the customer lifetime value. See equation (3)
-    from Fader et al. (2005) [1]_.
+    r"""Individual-level model for the customer lifetime value.
+
+    See equation (3) from Fader et al. (2005) [1]_.
 
     .. math::
 
@@ -100,15 +102,18 @@ class ContNonContract(PositiveContinuous):
     .. [1] Fader, Peter S., Bruce GS Hardie, and Ka Lok Lee. "“Counting your customers”
            the easy way: An alternative to the Pareto/NBD model." Marketing science
            24.2 (2005): 275-284.
+
     """
 
     rv_op = continuous_non_contractual
 
     @classmethod
     def dist(cls, lam, p, T, **kwargs):
+        """Get the distribution from the parameters."""
         return super().dist([lam, p, T], **kwargs)
 
     def logp(value, lam, p, T):
+        """Log-likelihood of the distribution."""
         t_x = value[..., 0]
         x = value[..., 1]
 
@@ -206,10 +211,10 @@ def _supp_shape_from_params(*args, **kwargs):
 
 
 class ContContract(PositiveContinuous):
-    r"""
-    Distribution class of a continuous contractual data-generating process,
-    that is where purchases can occur at any time point (continuous) and
-    churning/dropping out is explicit (contractual).
+    r"""Distribution class of a continuous contractual data-generating process.
+
+    That is where purchases can occur at any time point (continuous) and churning/dropping
+    out is explicit (contractual).
 
     .. math::
 
@@ -228,9 +233,11 @@ class ContContract(PositiveContinuous):
 
     @classmethod
     def dist(cls, lam, p, T, **kwargs):
+        """Get the distribution from the parameters."""
         return super().dist([lam, p, T], **kwargs)
 
     def logp(value, lam, p, T):
+        """Log-likelihood of the distribution."""
         t_x = value[..., 0]
         x = value[..., 1]
         churn = value[..., 2]
@@ -358,9 +365,9 @@ def _supp_shape_from_params(*args, **kwargs):
 
 
 class ParetoNBD(PositiveContinuous):
-    r"""
-    Population-level distribution class for a continuous, non-contractual, Pareto/NBD process,
-    based on Schmittlein, et al. in [2]_.
+    r"""Population-level distribution class for a continuous, non-contractual, Pareto/NBD process.
+
+    It is based on Schmittlein, et al. in [2]_.
 
     The likelihood function is derived from equations (22) and (23) of [3]_, with terms
     rearranged for numerical stability.
@@ -403,15 +410,18 @@ class ParetoNBD(PositiveContinuous):
     .. [3] Fader, Peter & G. S. Hardie, Bruce (2005).
            "A Note on Deriving the Pareto/NBD Model and Related Expressions."
            http://brucehardie.com/notes/009/pareto_nbd_derivations_2005-11-05.pdf
+
     """  # noqa: E501
 
     rv_op = pareto_nbd
 
     @classmethod
     def dist(cls, r, alpha, s, beta, T, **kwargs):
+        """Get the distribution from the parameters."""
         return super().dist([r, alpha, s, beta, T], **kwargs)
 
     def logp(value, r, alpha, s, beta, T):
+        """Log-likelihood of the distribution."""
         t_x = value[..., 0]
         x = value[..., 1]
 
@@ -555,9 +565,9 @@ def _supp_shape_from_params(*args, **kwargs):
 
 
 class BetaGeoBetaBinom(Discrete):
-    r"""
-    Population-level distribution class for a discrete, non-contractual, Beta-Geometric/Beta-Binomial process,
-    based on equation(5) from Fader, et al. in [1]_.
+    r"""Population-level distribution class for a discrete, non-contractual, Beta-Geometric/Beta-Binomial process.
+
+    It is based on equation(5) from Fader, et al. in [1]_.
 
     .. math::
 
@@ -584,9 +594,11 @@ class BetaGeoBetaBinom(Discrete):
 
     @classmethod
     def dist(cls, alpha, beta, gamma, delta, T, **kwargs):
+        """Get the distribution from the parameters."""
         return super().dist([alpha, beta, gamma, delta, T], **kwargs)
 
     def logp(value, alpha, beta, gamma, delta, T):
+        """Log-likelihood of the distribution."""
         t_x = pt.atleast_1d(value[..., 0])
         x = pt.atleast_1d(value[..., 1])
         scalar_case = t_x.type.broadcastable == (True,)

pymc_marketing/clv/__init__.py

@@ -11,6 +11,9 @@
 #   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 #   See the License for the specific language governing permissions and
 #   limitations under the License.
+
+"""CLV models."""
+
 from pymc_marketing.clv.models.basic import CLVModel
 from pymc_marketing.clv.models.beta_geo import BetaGeoModel
 from pymc_marketing.clv.models.gamma_gamma import (

pymc_marketing/clv/distributions.py

@@ -11,6 +11,8 @@
 #   WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 #   See the License for the specific language governing permissions and
 #   limitations under the License.
+"""CLV Model base class."""
+
 import json
 import warnings
 from collections.abc import Sequence
@@ -32,6 +34,8 @@
 
 
 class CLVModel(ModelBuilder):
+    """CLV Model base class."""
+
     _model_type = "CLVModel"
 
     @validate_call(config=ConfigDict(arbitrary_types_allowed=True))
@@ -69,6 +73,7 @@ def _validate_cols(
                     raise ValueError(f"Column {required_col} has duplicate entries")
 
     def __repr__(self) -> str:
+        """Representation of the model."""
         if not hasattr(self, "model"):
             return self._model_type
         else:
@@ -89,7 +94,7 @@ def fit(  # type: ignore
         fit_method: str = "mcmc",
         **kwargs,
     ) -> az.InferenceData:
-        """Infer model posterior
+        """Infer model posterior.
 
         Parameters
         ----------
@@ -99,8 +104,8 @@ def fit(  # type: ignore
             - "map": Finds maximum a posteriori via `pymc.find_MAP`
         kwargs:
             Other keyword arguments passed to the underlying PyMC routines
-        """
 
+        """
         self.build_model()  # type: ignore
 
         if fit_method == "mcmc":
@@ -120,8 +125,8 @@ def fit(  # type: ignore
         return self.idata
 
     def _fit_mcmc(self, **kwargs) -> az.InferenceData:
-        """
-        Fit a model using the data passed as a parameter.
+        """Fit a model using the data passed as a parameter.
+
         Sets attrs to inference data of the model.
 
 
@@ -138,6 +143,7 @@ def _fit_mcmc(self, **kwargs) -> az.InferenceData:
         -------
         self : az.InferenceData
             returns inference data of the fitted model.
+
         """
         sampler_config = {}
         if self.sampler_config is not None:
@@ -146,7 +152,7 @@ def _fit_mcmc(self, **kwargs) -> az.InferenceData:
         return pm.sample(**sampler_config, model=self.model)
 
     def _fit_MAP(self, **kwargs) -> az.InferenceData:
-        """Find model maximum a posteriori using scipy optimizer"""
+        """Find model maximum a posteriori using scipy optimizer."""
         model = self.model
         map_res = pm.find_MAP(model=model, **kwargs)
         # Filter non-value variables
@@ -162,8 +168,8 @@ def _fit_MAP(self, **kwargs) -> az.InferenceData:
 
     @classmethod
     def load(cls, fname: str):
-        """
-        Creates a ModelBuilder instance from a file,
+        """Create a ModelBuilder instance from a file.
+
         Loads inference data for the model.
 
         Parameters
@@ -179,12 +185,14 @@ def load(cls, fname: str):
         ------
         ValueError
             If the inference data that is loaded doesn't match with the model.
+
         Examples
         --------
         >>> class MyModel(ModelBuilder):
         >>>     ...
         >>> name = './mymodel.nc'
         >>> imported_model = MyModel.load(name)
+
         """
         filepath = Path(str(fname))
         idata = from_netcdf(filepath)
@@ -242,6 +250,7 @@ def thin_fit_result(self, keep_every: int):
 
     @property
     def default_sampler_config(self) -> dict:
+        """Default sampler configuration."""
         return {}
 
     @property
@@ -250,6 +259,7 @@ def _serializable_model_config(self) -> dict:
 
     @property
     def fit_result(self) -> Dataset:
+        """Get the fit result."""
         if self.idata is None or "posterior" not in self.idata:
             raise RuntimeError("The model hasn't been fit yet, call .fit() first")
         return self.idata["posterior"]
@@ -265,6 +275,7 @@ def fit_result(self, res: az.InferenceData) -> None:
             self.idata.posterior = res
 
     def fit_summary(self, **kwargs):
+        """Compute the summary of the fit result."""
         res = self.fit_result
         # Map fitting only gives one value, so we return it. We use arviz
         # just to get it nicely into a DataFrame
@@ -278,10 +289,13 @@ def fit_summary(self, **kwargs):
 
     @property
     def output_var(self):
+        """Output variable of the model."""
         pass
 
     def _generate_and_preprocess_model_data(self, *args, **kwargs):
+        """Generate and preprocess model data."""
         pass
 
     def _data_setter(self):
+        """Set the data for the model."""
         pass