from numpy import mean, max, min, multiply

def strategies(vector, strategy='average'):

''' Returns aggragated vector based on given startegy

Args:

vector (List[Float]): List of word vectors

strategy (str) : Strategy to aggregate the vector

Returns:

Array : Sentence Vector

'''

if strategy == 'average':

return mean(vector, axis=0)

elif strategy == 'min':

return min(vector)

elif strategy == 'max':

return max(vector)

else:

return multiply(vector)

def get_similarity(doc1, doc2, vectorizer):

''' Returns simlarity score of two documents

Args:

doc1 (str): Document to be compared

doc2 (str) : Document to be compared with

Returns:

Float : Similarity Score

'''

return vectorizer(doc1).similarity(vectorizer(doc2))

def get_annotation(sent,vectorizer):

''' Returns the Annotation of given senetence.This includes POS Tag

NER, Dependecy etc.

Args:

sent (str): sentence to be tagged

Retruns

spacy.token : Tagged iterator object.

'''

return vectorizer(sent)