Merge branch 'main' into cetagostini/multidimensionalmediamix

.github/release.yml

@@ -11,6 +11,9 @@ changelog:
     - title: Major Changes 🛠
       labels:
         - major
+    - title: Deprecations 🚨
+      labels:
+        - deprecation
     - title: New Features 🎉
       labels:
         - enhancement

docs/source/notebooks/mmm/mmm_budget_allocation_example.ipynb

@@ -1236,7 +1236,7 @@
     "initial_budget_scenario[\"t\"] = 0\n",
     "\n",
     "response_initial_budget = mmm.sample_posterior_predictive(\n",
-    "    X_pred=initial_budget_scenario, extend_idata=False\n",
+    "    initial_budget_scenario, extend_idata=False\n",
     ")\n",
     "\n",
     "response_initial_budget"

docs/source/notebooks/mmm/mmm_case_study.ipynb

@@ -6285,7 +6285,7 @@
    "outputs": [],
    "source": [
     "y_pred_test = mmm.sample_posterior_predictive(\n",
-    "    X_pred=X_test,\n",
+    "    X_test,\n",
     "    include_last_observations=True,\n",
     "    original_scale=True,\n",
     "    var_names=[\"y\", \"channel_contributions\"],\n",
@@ -6942,7 +6942,7 @@
     "    X_actual_uniform[control] = 0.0\n",
     "\n",
     "pred_test_uniform = mmm.sample_posterior_predictive(\n",
-    "    X_pred=X_actual_uniform,\n",
+    "    X_actual_uniform,\n",
     "    include_last_observations=True,\n",
     "    original_scale=True,\n",
     "    var_names=[\"y\", \"channel_contributions\"],\n",

docs/source/notebooks/mmm/mmm_counterfactuals.ipynb

@@ -701,7 +701,7 @@
    ],
    "source": [
     "y_forecast = mmm.sample_posterior_predictive(\n",
-    "    X_pred=X_forecast, extend_idata=False, include_last_observations=True\n",
+    "    X_forecast, extend_idata=False, include_last_observations=True\n",
     ")"
    ]
   },
@@ -958,7 +958,7 @@
    ],
    "source": [
     "y_intervention = mmm.sample_posterior_predictive(\n",
-    "    X_pred=X_intervention, extend_idata=False, include_last_observations=True\n",
+    "    X_intervention, extend_idata=False, include_last_observations=True\n",
     ")"
    ]
   },
@@ -1289,7 +1289,7 @@
    ],
    "source": [
     "y_counterfactual = mmm.sample_posterior_predictive(\n",
-    "    X_pred=X_counterfactual, extend_idata=False\n",
+    "    X_counterfactual, extend_idata=False\n",
     ");"
    ]
   },

docs/source/notebooks/mmm/mmm_example.ipynb

@@ -9661,7 +9661,7 @@
     "- `sample_posterior_predictive` : Get the full posterior predictive distribution\n",
     "- `predict`: Get the mean of the posterior predictive distribution\n",
     "\n",
-    "These methods take new data, `X_pred`, and some additional `kwargs` for new predictions. Namely, \n",
+    "These methods take new data, `X`, and some additional `kwargs` for new predictions. Namely, \n",
     "\n",
     "- `include_last_observations` : boolean flag in order to carry adstock effects from last observations in the training dataset"
    ]
@@ -10313,9 +10313,7 @@
     }
    ],
    "source": [
-    "y_out_of_sample = mmm.sample_posterior_predictive(\n",
-    "    X_pred=X_out_of_sample, extend_idata=False\n",
-    ")\n",
+    "y_out_of_sample = mmm.sample_posterior_predictive(X_out_of_sample, extend_idata=False)\n",
     "\n",
     "y_out_of_sample"
    ]
@@ -10435,7 +10433,7 @@
    ],
    "source": [
     "y_out_of_sample_with_adstock = mmm.sample_posterior_predictive(\n",
-    "    X_pred=X_out_of_sample, extend_idata=False, include_last_observations=True\n",
+    "    X_out_of_sample, extend_idata=False, include_last_observations=True\n",
     ")"
    ]
   },

docs/source/notebooks/mmm/mmm_time_slice_cross_validation.ipynb

@@ -260,7 +260,7 @@
     "    mmm = fit_mmm(mmm, X_train, y_train, random_seed)\n",
     "\n",
     "    y_pred_test = mmm.sample_posterior_predictive(\n",
-    "        X_pred=X_test,\n",
+    "        X_test,\n",
     "        include_last_observations=True,\n",
     "        original_scale=True,\n",
     "        extend_idata=False,\n",

docs/source/notebooks/mmm/mmm_tvp_example.ipynb

@@ -613,7 +613,7 @@
                 "    # Sample posterior predictive in whole data range (train and test)\n",
                 "    if \"posterior_predictive\" not in mmm.idata:\n",
                 "        mmm.sample_posterior_predictive(\n",
-                "            X_pred=DATA, extend_idata=True, var_names=[\"y\", \"intercept\"]\n",
+                "            DATA, extend_idata=True, var_names=[\"y\", \"intercept\"]\n",
                 "        )\n",
                 "    mmm.y = target_series.values\n",
                 "\n",

pymc_marketing/mlflow.py

@@ -673,7 +673,7 @@ class MMMWrapper(mlflow.pyfunc.PythonModel):
         Combine chain and draw dims into sample. Won't work if a dim named sample already exists. Defaults to True.
     include_last_observations : bool, default=False
         Boolean determining whether to include the last observations of the training data in order to carry over
-        costs with the adstock transformation. Assumes that X_pred are the next predictions following the
+        costs with the adstock transformation. Assumes that X are the next predictions following the
         training data. Defaults to False.
     original_scale : bool, default=True
         Boolean determining whether to return the predictions in the original scale of the target variable.
@@ -800,7 +800,7 @@ def log_mmm(
         already exists. Used for posterior/prior predictive sampling. Defaults to True.
     include_last_observations : bool, optional
         Whether to include the last observations of training data for adstock transformation.
-        Assumes X_pred are next predictions following training data. Used for all prediction
+        Assumes X are next predictions following training data. Used for all prediction
         methods. Defaults to False.
     original_scale : bool, optional
         Whether to return predictions in original scale of target variable. Used for all

pymc_marketing/mmm/hsgp.py

@@ -655,20 +655,19 @@ def parameterize_from_data(
         if isinstance(X, np.ndarray):
             numeric_X = np.asarray(X)
         elif isinstance(X, TensorVariable):
-            numeric_X = X.get_value(borrow=False)  # current numeric data
+            numeric_X = X.get_value(borrow=False)
         else:
             raise ValueError(
-                "X must be a NumPy array (or list) or a pm.Data/pm.MutableData. "
+                "X must be a NumPy array (or list) or a TensorVariable. "
                 "If it's a plain symbolic tensor, you must manually specify m, L."
             )
 
         numeric_X = np.asarray(numeric_X)
         if X_mid is None:
             X_mid = float(numeric_X.mean())
 
-        # 3. Use the standard approximation
         m, L = create_m_and_L_recommendations(
-            numeric_X,  # numeric version
+            numeric_X,
             X_mid,
             ls_lower=ls_lower,
             ls_upper=ls_upper,

pymc_marketing/mmm/mmm.py

@@ -55,6 +55,7 @@
     create_new_spend_data,
 )
 from pymc_marketing.mmm.validating import ValidateControlColumns
+from pymc_marketing.model_builder import _handle_deprecate_pred_argument
 from pymc_marketing.model_config import parse_model_config
 from pymc_marketing.prior import Prior
 
@@ -1902,7 +1903,7 @@ def legend_title_func(channel):
 
     def sample_posterior_predictive(
         self,
-        X_pred,
+        X=None,
         extend_idata: bool = True,
         combined: bool = True,
         include_last_observations: bool = False,
@@ -1913,15 +1914,15 @@ def sample_posterior_predictive(
 
         Parameters
         ----------
-        X_pred : array, shape (n_pred, n_features)
+        X : array, shape (n_pred, n_features)
             The input data used for prediction.
         extend_idata : bool, optional
             Boolean determining whether the predictions should be added to inference data object. Defaults to True.
         combined: bool, optional
             Combine chain and draw dims into sample. Won't work if a dim named sample already exists. Defaults to True.
         include_last_observations: bool, optional
             Boolean determining whether to include the last observations of the training data in order to carry over
-            costs with the adstock transformation. Assumes that X_pred are the next predictions following the
+            costs with the adstock transformation. Assumes that X are the next predictions following the
             training data.Defaults to False.
         original_scale: bool, optional
             Boolean determining whether to return the predictions in the original scale of the target variable.
@@ -1932,15 +1933,16 @@ def sample_posterior_predictive(
         Returns
         -------
         posterior_predictive_samples : DataArray, shape (n_pred, samples)
-            Posterior predictive samples for each input X_pred
+            Posterior predictive samples for each input X
 
         """
+        X = _handle_deprecate_pred_argument(X, "X", sample_posterior_predictive_kwargs)
         if include_last_observations:
-            X_pred = pd.concat(
-                [self.X.iloc[-self.adstock.l_max :, :], X_pred], axis=0
+            X = pd.concat(
+                [self.X.iloc[-self.adstock.l_max :, :], X], axis=0
             ).sort_values(by=self.date_column)
 
-        self._data_setter(X_pred)
+        self._data_setter(X)
 
         with self.model:  # sample with new input data
             post_pred = pm.sample_posterior_predictive(
@@ -2251,7 +2253,7 @@ def sample_response_distribution(
         constant_data = allocation_strategy.to_dataset(name="allocation")
 
         return self.sample_posterior_predictive(
-            X_pred=synth_dataset,
+            X=synth_dataset,
             extend_idata=False,
             include_last_observations=True,
             original_scale=False,