Add sentiment analysis example to neon

[nervanasys]

email from a user:

I am trying to get involved with text classification. I would like to start with the classical example of movie recomendations since there are a lot of examples using different kind of software to illustrate And solve the problem.

Vowpal wobbit, scikit learn, stanford nlp etc...

https://www.kaggle.com/c/sentiment-analysis-on-movie-reviews

I would also like to classify the typical sentiment140 dataset which contains tweet text Which can be noisy sometimes.

http://help.sentiment140.com/for-students/

[japerez20]

Hello guys: I am looking forward to learning more about neon. Do u have tutorial or slides where u show basic ideas of neon? Something like neon for dummies? Could you suggest me a link to understand better the Moby Dick example that you include with the first release of neon?

Thank, you, ja

[scttl]

Hi Ja,

Though we don't yet have a slide deck or screencast to share, I think the easiest way to get to grips with using neon is to start by skimming through the: quick start then following that up by looking at:

• neon/examples/mlp/mnist-small-noyaml.py to get a feel for the main objects and how they fit together
• neon/examples/ANNOTATED_EXAMPLE.yaml to understand how these pieces can be configured

As for the Moby Dick example, I'd suggest reading through the RNN docs as there's some detail in there describing how the data is laid out and so forth.

[yinyinl]

We have added the sentiment analysis example using movie review data. The example can be found at: imdb_lstm.py Along with the example script, we have added several layer components for word embeddings, processing recurrent layer outputs. Also some text preprocessing methods.