This is an intermediate issue, which is why it gets both the good first issue and good second issue tags.

We have an automated script to check when copies of the same code are consistent inside the library, which allows us to avoid subclassing and keep all code for one model's forward pass inside one file (see our philosophy for more details on this).

The XxxModelForYyy are very similar

Not a high-priority at all, but it'd be more sensible for such a tutorial/testing utility corpus to be implemented elsewhere - maybe under /test/ or some other data- or doc- related module – rather than in gensim.models.word2vec.

Originally posted by @gojomo in RaRe-Technologies/gensim#2939 (comment)

Unfortunately, I can't find any examples of what it "does" do, but it appears to take a file that is NOT in json format, but has a complete json representation on every single line of the file, with two keys: "text" and "label".

The code showing that is in its _read method:
....
def _read(self, file_path):
with open(cached_path(file_path), "r") as data_file:
for line in

Hi I would like to propose a better implementation for 'test_indices':

We can remove the unneeded np.array casting:

Cleaner/New:
test_indices = list(set(range(len(texts))) - set(train_indices))

Old:
test_indices = np.array(list(set(range(len(texts))) - set(train_indices)))