CODEWITHSUNDEEP
pythongensimdoc2vec
Shak
Shak

Reputation: 438

Type error when trying to create a doc2vec model in gensim

I am trying to train a Doc2Vec model using gensim.

The dataset i am using is the 20 newsgroups dataset [1] which is included in sklearn's datasets module.

I have used the example in the gensim documentation to create the model.

docs = newsgroups_train['data']
enumerated_docs = enumerate(docs)
documnets= [TaggedDocument(doc.split(),i) for i, doc in enumerated_docs]
model = Doc2Vec(documnets, vector_size=20, window=2, min_count=30, workers=4)

I checked every line of code, all seems to be working up to the line which initializes the model.

I get a type error: TypeError: 'int' object is not iterable

[1] https://scikit-learn.org/0.19/datasets/twenty_newsgroups.html

Upvotes: 0

Views: 255

Answers (1)

Michael Fengyuan Liu
Michael Fengyuan Liu

Reputation: 165

Enumerate returns an integer counter and the value in the list. So, in your third line of code, i is an integer. However, the second parameter of TaggedDocument function should be an iterable.

Upvotes: 1

Related Questions