model.py

import sys
import os
import cPickle

"""
This file provides an interface to 
a pre-trained politeness SVM. 
"""

#####
# Ensure the proper python dependencies exist

try:
    import numpy as np
except:
    sys.stderr.write("Package not found: Politeness model requires python package numpy\n")
    sys.exit(2)

try:
    import scipy
    from scipy.sparse import csr_matrix
except:
    sys.stderr.write("Package not found: Politeness model requires python package scipy\n")
    sys.exit(2)

try:
    import sklearn
except:
    sys.stderr.write("Package not found: Politeness model requires python package scikit-learn\n")
    sys.exit(2)

try:
    import nltk
except:
    sys.stderr.write("Package not found: Politeness model requires python package nltk\n")
    sys.exit(2)

####
# Check versions for sklearn, scipy, numpy, nltk
# Don't error out, just notify

packages2versions = [("scikit-learn", sklearn, "0.15.1"), ("numpy", np, "1.9.0"), ("nltk", nltk, "3.0.0"), ("scipy", scipy, "0.12.0")]

for name, package, expected_v in packages2versions:
    if package.__version__ < expected_v:
        sys.stderr.write("Warning: package '%s', expected version >= %s, detected %s. Code functionality not guaranteed.\n" % (name, expected_v, package.__version__))


####

from features.vectorizer import PolitenessFeatureVectorizer


####
# Serialized model filename

MODEL_FILENAME = os.path.join(os.path.split(__file__)[0], 'politeness-svm.p')

####
# Load model, initialize vectorizer

clf = cPickle.load(open(MODEL_FILENAME))
vectorizer = PolitenessFeatureVectorizer()

def score(request):
    """
    :param request - The request document to score
    :type request - dict with 'sentences' and 'parses' field
        sample (taken from test_documents.py)--
        {
            'sentences': [
                "Have you found the answer for your question?", 
                "If yes would you please share it?"
            ],
            'parses': [
                ["csubj(found-3, Have-1)", "dobj(Have-1, you-2)", "root(ROOT-0, found-3)", "det(answer-5, the-4)", "dobj(found-3, answer-5)", "poss(question-8, your-7)", "prep_for(found-3, question-8)"], 
                ["prep_if(would-3, yes-2)", "root(ROOT-0, would-3)", "nsubj(would-3, you-4)", "ccomp(would-3, please-5)", "nsubj(it-7, share-6)", "xcomp(please-5, it-7)"]
            ]
        } 

    returns class probabilities as a dict
        {
            'polite': float, 
            'impolite': float
        }
    """
    # vectorizer returns {feature-name: value} dict
    features = vectorizer.features(request)
    fv = [features[f] for f in sorted(features.iterkeys())]
    # Single-row sparse matrix
    X = csr_matrix(np.asarray([fv]))
    probs = clf.predict_proba(X)
    # Massage return format
    probs = {"polite": probs[0][1], "impolite": probs[0][0]}
    return probs



if __name__ == "__main__":

    """
    Sample classification of requests
    """

    from test_documents import TEST_DOCUMENTS

    for doc in TEST_DOCUMENTS:

        probs = score(doc)

        print "===================="
        print "Text: ", doc['text']
        print "\tP(polite) = %.3f" % probs['polite']
        print "\tP(impolite) = %.3f" % probs['impolite']
        print "\n"