[python] fix creating train_set in fit (#1916)

* [python] fix creating train_set in fit

https://github.com/Microsoft/LightGBM/blob/cc99f0d36ae929eb02b22a072823ab7c6d3155ab/python-package/lightgbm/sklearn.py#L519
may False even if valid_data[0] is X and valid_data[1] is y actually, because `check_X_y` might return copy of X and y.
https://scikit-learn.org/0.20/modules/generated/sklearn.utils.check_X_y.html

cf. https://github.com/Microsoft/LightGBM/pull/451

* use assertIn

python-package/lightgbm/sklearn.py

@@ -480,8 +480,10 @@ class LGBMModel(_LGBMModelBase):
             params['metric'] = set(original_metric + eval_metric)
 
         if not isinstance(X, DataFrame):
-            X, y = _LGBMCheckXY(X, y, accept_sparse=True, force_all_finite=False, ensure_min_samples=2)
-            _LGBMCheckConsistentLength(X, y, sample_weight)
+            _X, _y = _LGBMCheckXY(X, y, accept_sparse=True, force_all_finite=False, ensure_min_samples=2)
+            _LGBMCheckConsistentLength(_X, _y, sample_weight)
+        else:
+            _X, _y = X, y
 
         if self.class_weight is not None:
             class_sample_weight = _LGBMComputeSampleWeight(self.class_weight, y)
@@ -490,13 +492,13 @@ class LGBMModel(_LGBMModelBase):
             else:
                 sample_weight = np.multiply(sample_weight, class_sample_weight)
 
-        self._n_features = X.shape[1]
+        self._n_features = _X.shape[1]
 
         def _construct_dataset(X, y, sample_weight, init_score, group, params):
             ret = Dataset(X, label=y, weight=sample_weight, group=group, params=params)
             return ret.set_init_score(init_score)
 
-        train_set = _construct_dataset(X, y, sample_weight, init_score, group, params)
+        train_set = _construct_dataset(_X, _y, sample_weight, init_score, group, params)
 
         valid_sets = []
         if eval_set is not None:

tests/python_package_test/test_sklearn.py

@@ -279,3 +279,13 @@ class TestSklearn(unittest.TestCase):
         self.assertRaises(AssertionError,
                           np.testing.assert_allclose,
                           res_engine, res_sklearn_params)
+
+    def test_evaluate_train_set(self):
+        X, y = load_boston(True)
+        X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.1, random_state=42)
+        gbm = lgb.LGBMRegressor(n_estimators=10, silent=True)
+        gbm.fit(X_train, y_train, eval_set=[(X_train, y_train), (X_test, y_test)], verbose=False)
+        self.assertIn('training', gbm.evals_result_)
+        self.assertIn('l2', gbm.evals_result_['training'])
+        self.assertIn('valid_1', gbm.evals_result_)
+        self.assertIn('l2', gbm.evals_result_['valid_1'])