[python-package] free old dataset handle if set predictor/reference

python-package/lightgbm/basic.py

@@ -540,12 +540,19 @@ class Dataset(object):
         self.categorical_feature = categorical_feature
         self.params = params
         self.free_raw_data = free_raw_data
-        self._is_constructed = False
         self.used_indices = None
         self._predictor = None
 
     def __del__(self):
+        self._free_handle()
+
+    def _is_constructed(self):
+        return self.handle is not None
+
+    def _free_handle(self):
+        if self._is_constructed():
             _safe_call(_LIB.LGBM_DatasetFree(self.handle))
+        self.handle = None
 
     def _lazy_init(self, data, label=None, max_bin=255, reference=None,
                    weight=None, group=None, predictor=None,
@@ -587,7 +594,7 @@ class Dataset(object):
         """process for reference dataset"""
         ref_dataset = None
         if isinstance(reference, Dataset):
-            ref_dataset = reference.handle
+            ref_dataset = reference.construct().handle
         elif reference is not None:
             raise TypeError('Reference dataset should be None or dataset instance')
         """start construct data"""
@@ -718,8 +725,7 @@ class Dataset(object):
 
     def construct(self):
         """Lazy init"""
-        if not self._is_constructed:
-            self._is_constructed = True
+        if not self._is_constructed():
             if self.reference is not None:
                 if self.used_indices is None:
                     """create valid"""
@@ -729,10 +735,10 @@ class Dataset(object):
                 else:
                     """construct subset"""
                     used_indices = list_to_1d_numpy(self.used_indices, np.int32, name='used_indices')
-                    handle, self.handle = self.handle, ctypes.c_void_p()
+                    self.handle = ctypes.c_void_p()
                     params_str = param_dict_to_str(self.params)
                     _safe_call(_LIB.LGBM_DatasetGetSubset(
-                        handle,
+                        self.reference.construct().handle,
                         used_indices.ctypes.data_as(ctypes.POINTER(ctypes.c_int32)),
                         ctypes.c_int(used_indices.shape[0]),
                         c_str(params_str),
@@ -791,7 +797,6 @@ class Dataset(object):
                       categorical_feature=self.categorical_feature, params=params)
         ret._predictor = self._predictor
         ret.used_indices = used_indices
-        ret.handle = self.handle
         return ret
 
     def save_binary(self, filename):
@@ -824,6 +829,8 @@ class Dataset(object):
         data: numpy array or list or None
             The array ofdata to be set
         """
+        if not self._is_constructed():
+            raise Exception("cannot set filed before construct dataset handle")
         if data is None:
             """set to None"""
             _safe_call(_LIB.LGBM_DatasetSetField(
@@ -865,6 +872,8 @@ class Dataset(object):
         info : array
             A numpy array of information of the data
         """
+        if not self._is_constructed():
+            raise Exception("cannot Get filed before construct dataset handle")
         tmp_out_len = ctypes.c_int()
         out_type = ctypes.c_int()
         ret = ctypes.POINTER(ctypes.c_void_p)()
@@ -899,7 +908,7 @@ class Dataset(object):
             return
         if self.data is not None:
             self.categorical_feature = categorical_feature
-            self._is_constructed = False
+            self._free_handle()
         else:
             raise LightGBMError("Cannot set categorical feature after freed raw data,\
              Set free_raw_data=False when construct Dataset to avoid this.")
@@ -913,7 +922,7 @@ class Dataset(object):
             return
         if self.data is not None:
             self._predictor = predictor
-            self._is_constructed = False
+            self._free_handle()
         else:
             raise LightGBMError("Cannot set predictor after freed raw data,Set free_raw_data=False when construct Dataset to avoid this.")
 
@@ -933,7 +942,7 @@ class Dataset(object):
             return
         if self.data is not None:
             self.reference = reference
-            self._is_constructed = False
+            self._free_handle()
         else:
             raise LightGBMError("Cannot set reference after freed raw data,\
              Set free_raw_data=False when construct Dataset to avoid this.")
@@ -948,7 +957,7 @@ class Dataset(object):
             Feature names
         """
         self.feature_name = feature_name
-        if self._is_constructed and feature_name is not None:
+        if self._is_constructed() and feature_name is not None:
             if len(feature_name) != self.num_feature():
                 raise ValueError("Length of feature_name({}) and num_feature({}) don't match".format(len(feature_name), self.num_feature()))
             c_feature_name = [c_str(name) for name in feature_name]
@@ -967,7 +976,7 @@ class Dataset(object):
             The label information to be set into Dataset
         """
         self.label = label
-        if self._is_constructed:
+        if self._is_constructed():
             label = list_to_1d_numpy(label, name='label')
             self.set_field('label', label)
 
@@ -981,7 +990,7 @@ class Dataset(object):
             Weight for each data point
         """
         self.weight = weight
-        if self._is_constructed and weight is not None:
+        if self._is_constructed() and weight is not None:
             weight = list_to_1d_numpy(weight, name='weight')
             self.set_field('weight', weight)
 
@@ -995,7 +1004,7 @@ class Dataset(object):
             Init score for booster
         """
         self.init_score = init_score
-        if self._is_constructed and init_score is not None:
+        if self._is_constructed() and init_score is not None:
             init_score = list_to_1d_numpy(init_score, name='init_score')
             self.set_field('init_score', init_score)
 
@@ -1009,7 +1018,7 @@ class Dataset(object):
             Group size of each group
         """
         self.group = group
-        if self._is_constructed and group is not None:
+        if self._is_constructed() and group is not None:
             group = list_to_1d_numpy(group, np.int32, name='group')
             self.set_field('group', group)
 
@@ -1021,7 +1030,7 @@ class Dataset(object):
         -------
         label : array
         """
-        if self.label is None and self._is_constructed:
+        if self.label is None and self._is_constructed():
             self.label = self.get_field('label')
         return self.label
 
@@ -1033,7 +1042,7 @@ class Dataset(object):
         -------
         weight : array
         """
-        if self.weight is None and self._is_constructed:
+        if self.weight is None and self._is_constructed():
             self.weight = self.get_field('weight')
         return self.weight
 
@@ -1045,7 +1054,7 @@ class Dataset(object):
         -------
         init_score : array
         """
-        if self.init_score is None and self._is_constructed:
+        if self.init_score is None and self._is_constructed():
             self.init_score = self.get_field('init_score')
         return self.init_score
 
@@ -1057,7 +1066,7 @@ class Dataset(object):
         -------
         init_score : array
         """
-        if self.group is None and self._is_constructed:
+        if self.group is None and self._is_constructed():
             self.group = self.get_field('group')
             if self.group is not None:
                 # group data from LightGBM is boundaries data, need to convert to group size
@@ -1075,7 +1084,7 @@ class Dataset(object):
         -------
         number of rows : int
         """
-        if self._is_constructed:
+        if self._is_constructed():
             ret = ctypes.c_int()
             _safe_call(_LIB.LGBM_DatasetGetNumData(self.handle,
                                                    ctypes.byref(ret)))
@@ -1091,7 +1100,7 @@ class Dataset(object):
         -------
         number of columns : int
         """
-        if self._is_constructed:
+        if self._is_constructed():
             ret = ctypes.c_int()
             _safe_call(_LIB.LGBM_DatasetGetNumFeature(self.handle,
                                                       ctypes.byref(ret)))