Why doesn't gensim Word2Vec recognize the compute_loss keyword?

According to the gensim.models.Word2Vec API reference , "compute_loss" is a valid keyword. However, I am getting an error unexpected keyword

.

UPDATE

The Word2Vec class on GitHub has the 'compute_loss' keyword, but my local library does not. I can see that the documentation and gensim library are diverging from each other. I found that the win-64/gensim-2.2.0-np113py35_0.tar.bz2

file in the conda repository is not updated.

However, after removing gensim from conda, pip install gensim

didn't change anything as it still doesn't work.

Apparently the source on GitHub and the distributed library are different, but the tutorial seems to assume the code is the same as on GitHub.

/ END UPDATE

I followed and downloaded the tutorial on Word2Vec .

In the input field [25], the first cell after the heading "Learning Loss", I get an error in the Word2Vec class' initializer.

Input:

# instantiating and training the Word2Vec model
model_with_loss = gensim.models.Word2Vec(sentences, min_count=1, 
compute_loss=True, hs=0, sg=1, seed=42)

# getting the training loss value
training_loss = model_with_loss.get_latest_training_loss()
print(training_loss)

      

        
        
        
      

    

Output:

---------------------------------------------------------------------------
TypeError                                 Traceback (most recent call last)
<ipython-input-25-c2933abf4b08> in <module>()
      1 # instantiating and training the Word2Vec model
----> 2 model_with_loss = gensim.models.Word2Vec(sentences, min_count=1, compute_loss=True, hs=0, sg=1, seed=42)
      3 
      4 # getting the training loss value
      5 training_loss = model_with_loss.get_latest_training_loss()

TypeError: __init__() got an unexpected keyword argument 'compute_loss'

      

        
        
        
      

    

I have gensim 2.2.0 installed via conda and a fresh new clone from the gensim repository (with a study laptop). I am using 64-bit Python 3.5.3 on Windows 10. (Anaconda)

I tried to look for others with the same encounter, but I was not successful.

Do you know the reason for this and how to fix it? Apparently the source on GitHub and the distributed library are different, but the tutorial seems to assume the code is the same as on GitHub.

I also posted a question on the official mailing list earlier .