Merge pull request #58 from rodrigo-arenas/0.6.X

Release notes

dev-requirements.txt

@@ -6,7 +6,7 @@ twine==3.3.0
 numpy>=1.13.3
 seaborn>=0.11.1
 deap>=1.3.1
-mlflow==1.17.0
+mlflow>=1.17.0
 black==21.5b2
 sphinx
 sphinx_gallery

docs/release_notes.rst

@@ -22,12 +22,12 @@ Features:
 * Added new parallel coordinates plot in  :func:`~sklearn_genetic.plots.plot_parallel_coordinates`.
 * Now if one or more callbacks decides to stop the algorithm, it will print
   its class name to know which callbacks were responsible of the stopping.
-* Added support for extra methods coming from scikit-learn's BaseSearchCV, it is
-  still partial support, missing properties like `cv_results_`, `best_index_` and `multimetric_`.
+* Added support for extra methods coming from scikit-learn's BaseSearchCV, like `cv_results_`,
+  `best_index_` and `refit_time_` among others.
 * Added methods `on_start` and `on_end` to :class:`~sklearn_genetic.callbacks.base.BaseCallback`.
   Now the algorithms check for the callbacks like this:
 
-  - **on_start**: When the evolutionary algorithm is called from the GASearchCV.fit method
+  - **on_start**: When the evolutionary algorithm is called from the GASearchCV.fit method.
 
   - **on_step:** When the evolutionary algorithm finish a generation (no change here).
 
@@ -48,7 +48,9 @@ API Changes:
   now requires an explicit installation of seaborn and mlflow, now those
   are optionally installed using ``pip install sklearn-genetic-opt[all].``
 * The GASearchCV.logbook property now has extra information that comes from the
-  scikit-learn cross_validate function
+  scikit-learn cross_validate function.
+* An optional extra parameter was added to GASearchCV, named `return_train_score`: bool, default= ``False``.
+  As in scikit-learn, it controls if the `cv_results_` should have the training scores.
 
 ^^^^^
 Docs:

sklearn_genetic/callbacks/base.py

@@ -54,6 +54,7 @@ def on_end(self, logbook=None, estimator=None):
 
         """
         pass  # pragma: no cover
-
+    
     def __call__(self, record=None, logbook=None, estimator=None):
         return self.on_step(record, logbook, estimator)
+

sklearn_genetic/genetic_search.py

@@ -441,6 +441,7 @@ def fit(self, X, y, callbacks=None):
 
         # Make sure the callbacks are valid
         self.callbacks = check_callback(callbacks)
+
         self.scorer_ = check_scoring(self.estimator, scoring=self.scoring)
 
         # Check cv and get the n_splits
@@ -481,16 +482,15 @@ def fit(self, X, y, callbacks=None):
             self.estimator.set_params(**self.best_params_)
 
             refit_start_time = time.time()
-
             self.estimator.fit(self.X_, self.y_)
             refit_end_time = time.time()
             self.refit_time_ = refit_end_time - refit_start_time
 
             self.best_estimator_ = self.estimator
 
             # hof keeps the best params according to the fitness value
-            # To be consistent with self.best_estimator_, if more than 1 model gets same score
-            # It could lead to differences between hof and self.best_estimator_
+            # To be consistent with self.best_estimator_, if more than 1 model gets the
+            # same score, it could lead to differences between hof and self.best_estimator_
             self._hof.remove(0)
             self._hof.items.insert(0, list(self.best_params_.values()))
             self._hof.keys.insert(0, self.best_score_)