Balaka Biswas

@jawsvk make sure your PR comes in on 1st October, because Code Coverage is vital before the code based PRs start pouring in.

> Is it okay if I work on this issue during Hacktoberfest? Yes sure @2bit-hack . Just make sure the PR is made after Oct 1. Assigning it to you.

> I have a few questions regarding the implementation: > > * Will there be a corpus of documents provided against which to run the TF-IDF algorithm? > (Something like...

> tf-idf is a corpus based algorithm whereas RAKE can work on single documents. Like, if I have a collection of documents, I can figure out the tf-idf scores for...

Oh yes..thanks for pointing that out. Yes you change the function to include the corpus. Make sure you include the ```max_num``` parameter.

> The code for keyword extraction using TF-IDF goes into `tfidf_vectorizer/extract_keywords_tfidf_scratch.py`. Where do the tests of RAKE vs TF-IDF go? I noticed there's a `tests` folder, does it go there?...

@2bit-hack Basically you have to toil a bit. What they have done in the paper is that, they have marked out keywords manually. Then they ran the algorithms, and checked...

@johnnthough the black one's better. Also, resolve the branch conflicts.