def get_answer(question):
"""
:param question: String with the question.
:return: Filled Answer object.
"""
a = Answer(Question(question))
checked = check_keywords(a.q)
if not checked:
a.text = 'Didn\'t understand the question'
return a
# print warning on weird keywords
# (XXX: should we give up on these questions instead?)
check_unknowns(a)
if len(a.q.unknown) > 0:
print 'we have no information on these words:', a.q.unknown
found_sources, found_anything = get_content_elastic(a, search_all=False)
if found_sources:
gen_features(a)
a.text = answer_all(a)
return a
if not found_anything:
a.text = 'Absolutely no result'
else:
a.text = 'No result'
return a
def get_answer(question):
"""
:param question: String with the question.
:return: Filled Answer object.
note: if question starts with '>>>', just ask elasticsearch
"""
if question.startswith('>>>'):
return ask_only(question[3:])
a = Answer(Question(question))
checked = check_keywords(a.q)
if not checked:
a.q.query += ' (' + ','.join(a.q.not_in_kw) + ' not in keywords)'
a.text = 'Didn\'t understand the question'
return a
# print warning on weird keywords
# (XXX: should we give up on these questions instead?)
check_unknowns(a)
if len(a.q.unknown) > 0:
print 'we have no information on these words:', a.q.unknown
found_sources, found_anything = get_content_elastic(a, search_all=False)
if found_sources:
gen_features(a)
a.text = answer_all(a)
return a
if not found_anything:
a.text = 'Absolutely no result'
else:
a.text = 'No result'
return a
def get_answer(question):
"""
:param question: String with the question.
:return: Filled Answer object.
note: if question starts with '>>>', just ask elasticsearch
"""
if question.startswith('>>>'):
return ask_only(question[3:])
a = Answer(Question(question))
checked = check_keywords(a.q)
if not checked:
a.q.query += ' (' + ','.join(a.q.not_in_kw) + ' not in keywords)'
a.text = 'Didn\'t understand the question'
return a
# print warning on weird keywords
# (XXX: should we give up on these questions instead?)
check_unknowns(a)
if len(a.q.unknown) > 0:
print 'we have no information on these words:', a.q.unknown
found_sources, found_anything = get_content_elastic(a, search_all=False)
if found_sources:
gen_features(a)
a.text = answer_all(a)
return a
if not found_anything:
a.text = 'Absolutely no result'
else:
a.text = 'No result'
return a
def ask_only(query):
a = Answer(Question(preprocess_question(query)))
check_unknowns(a)
if len(a.q.unknown) > 0:
print 'we have no information on these words:', a.q.unknown
found_sources, found_anything = ask(a, query)
if found_sources:
gen_features(a)
a.text = answer_all(a)
a.text = 'Query only'
return a
if not found_anything:
a.text = 'Absolutely no result'
return a
def ask_only(query):
a = Answer(Question(preprocess_question(query)))
check_unknowns(a)
if len(a.q.unknown) > 0:
print 'we have no information on these words:', a.q.unknown
found_sources, found_anything = ask(a, query)
if found_sources:
gen_features(a)
a.text = answer_all(a)
a.text = 'Query only'
return a
if not found_anything:
a.text = 'Absolutely no result'
return a
def main():
n = 12
m = 2
if len(sys.argv) <= 1:
print('Usage is main [gen|run] [seed]')
return
elif len(sys.argv) >= 2:
if len(sys.argv) >= 3:
random.seed(int(sys.argv[2]))
A, B, a = gen_features(n, m)
C = [[A[i][0], A[i][1], B[i][0], B[i][1]] for i in range(len(A))]
if sys.argv[1] == 'gen':
C.insert(0, ['A-Radius', 'A-Perimeter', 'B-Radius', 'B-Perimeter'])
tableutils.printmat(C, True)
elif sys.argv[1] == 'plane':
a = [a]
a.insert(0, ['a1', 'a2', 'a0'])
tableutils.printmat(a)
elif sys.argv[1] == 'run':
rc, Value, G = solve_classification(A, B)
T = [[[rc, Value], G]]
tableutils.printmat(T)
elif sys.argv[1] == 'margins':
rc, a = solve_margins_classification(A, B)
if rc == 0:
low = min([A[i][0] for i in range(len(A))] +
[B[i][0] for i in range(len(B))])
high = 1 + max([A[i][0] for i in range(len(A))] +
[B[i][0] for i in range(len(B))])
T = [[x, a[2] / a[1] - a[0] / a[1] * x]
for x in range(low, high)]
else:
T = [['Infeasible']]
tableutils.printmat(T, True)
def classify(name):
name = name.upper()
if name in cache_dict:
return cache_dict[name]
else:
label = classifier.classify(gen_features(name))
if cache:
cache_dict[name] = label
return label
def classify(name):
name = name.upper()
if name in cache_dict:
return cache_dict[name]
else:
feature = gen_features(name)
label_numeric = classifier.predict(feature)[0]
label = classes[label_numeric]
if cache:
cache_dict[name] = label
return label
def test_compound_surnames():
name = "Adi du Preez"
name_features = features.gen_features(name,
include_name_part=False,
include_bigrams=False,
include_trigrams=False,
include_4grams=False,
include_surname=True,
include_fullname=False)
ngrams = ["surname_DU_PREEZ"]
for ngram in ngrams:
assert ngram in name_features
assert len(name_features) == len(ngrams)
def test_gen_features():
name = "Adi Eyal"
name_features = features.gen_features(name,
include_name_part=False,
include_bigrams=False,
include_trigrams=False,
include_4grams=False,
include_surname=False,
include_fullname=False)
assert len(name_features) == 0
name_features = features.gen_features(name,
include_name_part=False,
include_bigrams=False,
include_trigrams=False,
include_4grams=False,
include_surname=False,
include_fullname=True)
assert len(name_features) == 1
assert "name_ADI_EYAL" in name_features
name_features = features.gen_features(name,
include_name_part=False,
include_bigrams=False,
include_trigrams=False,
include_4grams=False,
include_surname=True,
include_fullname=False)
name_features = features.gen_features(name,
include_name_part=False,
include_bigrams=False,
include_trigrams=False,
include_4grams=False,
include_surname=True,
include_fullname=False)
assert len(name_features) == 1
assert "surname_EYAL" in name_features
name_features = features.gen_features(name,
include_name_part=True,
include_bigrams=False,
include_trigrams=False,
include_4grams=False,
include_surname=False,
include_fullname=False)
assert len(name_features) == 2
assert "namepart_ADI" in name_features
assert "namepart_EYAL" in name_features
name_features = features.gen_features(name,
include_name_part=False,
include_bigrams=True,
include_trigrams=False,
include_4grams=False,
include_surname=False,
include_fullname=False)
assert len(name_features) == 9
ngrams = ["AD", "DI", "EY", "YA", "AL", "b_AD", "b_EY", "l_DI", "l_AL"]
for ngram in ngrams:
assert ngram in name_features
assert len(name_features) == len(ngrams)
name_features = features.gen_features(name,
include_name_part=False,
include_bigrams=False,
include_trigrams=True,
include_4grams=False,
include_surname=False,
include_fullname=False)
assert len(name_features) == 7
ngrams = ["ADI", "EYA", "YAL", "b_ADI", "l_ADI", "b_EYA", "l_YAL"]
for ngram in ngrams:
assert ngram in name_features
assert len(name_features) == len(ngrams)
name_features = features.gen_features(name,
include_name_part=False,
include_bigrams=False,
include_trigrams=False,
include_4grams=True,
include_surname=False,
include_fullname=False)
assert len(name_features) == 3
ngrams = ["EYAL", "b_EYAL", "l_EYAL"]
for ngram in ngrams:
assert ngram in name_features
assert len(name_features) == len(ngrams)
def classify_many(names):
features = [gen_features(name) for name in names]
labels = classifier.predict(features)
return [classes[l] for l in labels]