Gensim: Add ELMo (Deep contextualized word representations)

[chengrufeng]

Introduction

ELMo is a deep contextualized word representation that models both (1) complex characteristics of word use (e.g., syntax and semantics), and (2) how these uses vary across linguistic contexts (i.e., to model polysemy). These word vectors are learned functions of the internal states of a deep bidirectional language model (biLM), which is pre-trained on a large text corpus. They can be easily added to existing models and significantly improve the state of the art across a broad range of challenging NLP problems, including question answering, textual entailment and sentiment analysis.

Other Tools

• [ ] [allenai](https://github.com/allenai/bilm-tf)
• [ ] [elmo](https://allennlp.org/elmo)

Issue Description

It's not easy to use them.
I need a more simple and convenient tool like "Word2Vec" in "gensim"

[qichaotang]

attention

[menshikh-iv]

Sounds promising for me, but need deep investigation first (i.e. reproduce experiments, see how it will work by quality/performance/etc).

@piskvorky @gojomo wdyt?

[justinas-kazanavicius]

Any plans for this?

[menshikh-iv]

@justinas-kazanavicius not yet, gensim now in "slow maintenance mode", I don't think we'll have a time for this feature. BTW, other packages already implement that, what's a reason to make "another yet implementation"?