DenseBoW.select bug

EvoMSA/__init__.py

@@ -11,7 +11,7 @@
 # WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
 # See the License for the specific language governing permissions and
 # limitations under the License.
-__version__ = '1.9.9'
+__version__ = '1.9.10'
 
 try:
     from EvoMSA.text_repr import BoW, TextRepresentations, StackGeneralization, DenseBoW

EvoMSA/tests/test_text_repr.py

@@ -449,4 +449,22 @@ def test_DenseBoW_extend2():
 def test_DenseBoW_dataset():
     from EvoMSA.text_repr import DenseBoW
     dense = DenseBoW(lang='it', emoji=False, keyword=False)
-
+
+
+
+def test_DenseBoW_select_bug():
+    from EvoMSA.text_repr import DenseBoW
+    from EvoMSA.utils import MICROTC
+    D = list(tweet_iterator(TWEETS))
+    pos = [x for x in D if x['klass'] == 'P']
+    neg = [x for x in D if x['klass'] == 'N']
+    lang = 'es'
+    name = 'emojis'
+    func = 'most_common_by_type'
+    d = 13
+    text_repr = DenseBoW(lang=lang,
+                         keyword=False,
+                         voc_size_exponent=13,
+                         emoji=True, dataset=False,
+                         n_jobs=-1)
+    text_repr.select(D=pos + neg[:1]).fit(D)

EvoMSA/text_repr.py

@@ -678,6 +678,8 @@ def select(self, subset: Union[list, None]=None,
         """
         assert subset is not None or D is not None
         if subset is not None:
+            if len(subset) == 0:
+                return self
             tr = self.text_representations
             self.text_representations = [tr[i] for i in subset]
             names = self.names