Throw exception if load() is called on instance rather than the class…

… in word2vec and doc2vec (#889)

* Update word2vec.py

* Update word2vec.py

* Update word2vec.py

* Update doc2vec.py

* Update doc2vec.py

* Update word2vec.py

* Update utils.py

* Update word2vec.py

* Update utils.py

* Update word2vec.py

* Update doc2vec.py

* Update doc2vec.py

* Update word2vec.py

* Update utils.py

* Update word2vec.py

* Update doc2vec.py

* Update doc2vec.py

* Update utils.py

* 	modified:   gensim/utils.py

* 	modified:   gensim/test/test_doc2vec.py
	modified:   gensim/test/test_word2vec.py

* 	modified:   gensim/test/test_word2vec.py

* 	modified:   gensim/test/test_doc2vec.py
	modified:   gensim/test/test_word2vec.py

* 	modified:   gensim/test/test_doc2vec.py
	modified:   gensim/test/test_word2vec.py

* 	modified:   gensim/test/test_doc2vec.py
	modified:   gensim/test/test_word2vec.py

* 	modified:   gensim/test/test_doc2vec.py
	modified:   gensim/test/test_word2vec.py

* 	modified:   gensim/test/test_doc2vec.py
	modified:   gensim/test/test_word2vec.py

* Update test_word2vec.py

* Update test_doc2vec.py

* Update test_doc2vec.py

* Update test_word2vec.py

* Update doc2vec.py

* Update doc2vec.py

* Update test_word2vec.py

* Update test_word2vec.py

* Update test_word2vec.py

* Update test_doc2vec.py

* Update test_word2vec.py

gensim/models/doc2vec.py

@@ -50,6 +50,8 @@
     repeat as np_repeat, array, float32 as REAL, empty, ones, memmap as np_memmap, \
     sqrt, newaxis, ndarray, dot, vstack, dtype, divide as np_divide
 
+
+from gensim.utils import call_on_class_only
 from gensim import utils, matutils  # utility fnc for pickling, common scipy operations etc
 from gensim.models.word2vec import Word2Vec, Vocab, train_cbow_pair, train_sg_pair, train_batch_sg
 from six.moves import xrange, zip
@@ -603,10 +605,13 @@ def __init__(self, documents=None, dm_mean=None,
         super(Doc2Vec, self).__init__(
             sg=(1 + dm) % 2,
             null_word=dm_concat, **kwargs)
+
+        self.load = call_on_class_only
+        self.load_word2vec_format = call_on_class_only
 
         if dm_mean is not None:
             self.cbow_mean = dm_mean
-
+        
         self.dbow_words = dbow_words
         self.dm_concat = dm_concat
         self.dm_tag_count = dm_tag_count

gensim/models/word2vec.py

@@ -78,6 +78,7 @@
 import threading
 import itertools
 
+from gensim.utils import keep_vocab_item, call_on_class_only
 from gensim.utils import keep_vocab_item
 from gensim.models.keyedvectors import KeyedVectors
 
@@ -421,6 +422,9 @@ def __init__(
 
         """
 
+        self.load = call_on_class_only
+        self.load_word2vec_format = call_on_class_only        
+
         if FAST_VERSION == -1:
             logger.warning('Slow version of {0} is being used'.format(__name__))
         else:

gensim/test/test_doc2vec.py

@@ -62,6 +62,12 @@ def testfile():
     # temporary data will be stored to this file
     return os.path.join(tempfile.gettempdir(), 'gensim_doc2vec.tst')
 
+def load_on_instance():
+    # Save and load a Doc2Vec Model on instance for test
+    model = doc2vec.Doc2Vec(DocsLeeCorpus(), min_count=1)
+    model.save(testfile())
+    model = doc2vec.Doc2Vec() # should fail at this point
+    return model.load(testfile())
 
 class TestDoc2VecModel(unittest.TestCase):
     def test_persistence(self):
@@ -342,6 +348,10 @@ def testTrainWarning(self, l):
                 model.alpha += 0.05
         warning = "Effective 'alpha' higher than previous training cycles"
         self.assertTrue(warning in str(l))
+
+    def testLoadOnClassError(self):
+        """Test if exception is raised when loading doc2vec model on instance"""
+        self.assertRaises(AttributeError, load_on_instance)
 #endclass TestDoc2VecModel
 
 

gensim/test/test_word2vec.py

@@ -70,13 +70,17 @@ def testfile():
     # temporary data will be stored to this file
     return os.path.join(tempfile.gettempdir(), 'gensim_word2vec.tst')
 
-
 def _rule(word, count, min_count):
     if word == "human":
         return utils.RULE_DISCARD  # throw out
     else:
         return utils.RULE_DEFAULT  # apply default rule, i.e. min_count
-
+def load_on_instance():
+    # Save and load a Word2Vec Model on instance for test
+    model = word2vec.Word2Vec(sentences, min_count=1)
+    model.save(testfile())
+    model = word2vec.Word2Vec() # should fail at this point
+    return model.load(testfile())
 
 class TestWord2VecModel(unittest.TestCase):
     def testOnlineLearning(self):
@@ -585,14 +589,18 @@ def testTrainWarning(self, l):
                 model.alpha += 0.05
         warning = "Effective 'alpha' higher than previous training cycles"
         self.assertTrue(warning in str(l))
-#endclass TestWord2VecModel
-
+
     def test_sentences_should_not_be_a_generator(self):
         """
         Is sentences a generator object?
         """
         gen = (s for s in sentences)
         self.assertRaises(TypeError, word2vec.Word2Vec, (gen,))
+
+    def testLoadOnClassError(self):
+        """Test if exception is raised when loading word2vec model on instance"""
+        self.assertRaises(AttributeError, load_on_instance)
+#endclass TestWord2VecModel
 
 class TestWMD(unittest.TestCase):
     def testNonzero(self):

gensim/utils.py

@@ -216,6 +216,10 @@ def any2unicode(text, encoding='utf8', errors='strict'):
     return unicode(text, encoding, errors=errors)
 to_unicode = any2unicode
 
+def call_on_class_only(*args, **kwargs):
+    """Raise exception when load methods are called on instance"""
+    raise AttributeError('This method should be called on a class object.')
+
 
 class SaveLoad(object):
     """