Gets the 4096-dimensional CNN features for the given COCO images Input: img_coco_ids: A list of strings, each string corresponding to t

Returns a time series of word vectors for tokens in the question Input: questions: list of unicode objects nlp: an instance of the class English

Converts string objects to class labels Input: answers: a list of unicode objects encoder: a scikit-learn LabelEncoder object Output: A numpy

Sums the word vectors of all the tokens in a question Input: questions: list of unicode objects nlp: an instance of the class English() from sp

Before runnning this demo ensure that you have some images from the MS COCO validation set saved somewhere, and update the image_dir variable accor