fix_gridsearch

Author: thomas0125

libmultilabel/linear/utils.py

@@ -109,32 +109,31 @@ class GridSearchCV(sklearn.model_selection.GridSearchCV):
     The usage is similar to sklearn's, except that the parameter ``scoring`` is unavailable. Instead, specify ``scoring_metric`` in ``MultiLabelEstimator`` in the Pipeline.
 
     Args:
-        pipeline (sklearn.pipeline.Pipeline): A sklearn Pipeline for grid search.
+        estimator (estimator object): A estimator for grid search.
         param_grid (dict): Search space for a grid search containing a dictionary of
             parameters and their corresponding list of candidate values.
         n_jobs (int, optional): Number of CPU cores run in parallel. Defaults to None.
     """
 
-    _required_parameters = ["pipeline", "param_grid"]
+    _required_parameters = ["estimator", "param_grid"]
 
-    def __init__(self, pipeline: sklearn.pipeline.Pipeline, param_grid: dict, n_jobs=None, **kwargs):
-        assert isinstance(pipeline, sklearn.pipeline.Pipeline)
+    def __init__(self, estimator, param_grid: dict, n_jobs=None, **kwargs):
         if n_jobs is not None and n_jobs > 1:
-            param_grid = self._set_singlecore_options(pipeline, param_grid)
+            param_grid = self._set_singlecore_options(estimator, param_grid)
         if "scoring" in kwargs.keys():
             raise ValueError(
                 "Please specify the validation metric with `MultiLabelEstimator.scoring_metric` in the Pipeline instead of using the parameter `scoring`."
             )
 
-        super().__init__(estimator=pipeline, n_jobs=n_jobs, param_grid=param_grid, **kwargs)
+        super().__init__(estimator=estimator, n_jobs=n_jobs, param_grid=param_grid, **kwargs)
 
-    def _set_singlecore_options(self, pipeline: sklearn.pipeline.Pipeline, param_grid: dict):
+    def _set_singlecore_options(self, estimator, param_grid: dict):
         """Set liblinear options to `-m 1`. The grid search option `n_jobs`
         runs multiple processes in parallel. Using multithreaded liblinear
         in conjunction with grid search oversubscribes the CPU and deteriorates
         the performance significantly.
         """
-        params = pipeline.get_params()
+        params = estimator.get_params()
         for name, transform in params.items():
             if isinstance(transform, MultiLabelEstimator):
                 regex = r"-m \d+"