else:
continue
if ( m_score > 0.0 ):
machine_score.append(m_score)
human_score.append(float(p[2]))
p_val, p_rel = sci.stats.spearmanr(human_score, machine_score)
print "Simple Linear Approach", p_val
if __name__ == "__main__":
word_vectors = nlp.read_word_vectors(VECTOR_DIR + VECTOR_NAME)
word_pairs = nlp.read_csv(CSV_DIR + CSV_NAME)
vocab = []
for p in word_pairs:
vocab.append(p[0].lower())
vocab.append(p[1].lower())
vocab = list(set(vocab))
for w in vocab:
word_hypernyms[w] = nlp.read_hypernyms(w)
word_hyponyms[w] = nlp.read_hyponyms(w)
word_synonyms[w] = nlp.read_synonyms(w)
senses[w] = nlp.read_senses(w)
for s in senses[w]:
sense_vectors[s] = np.zeros(VECTOR_DIM)
sense_hypernyms[s] = nlp.read_hypernyms_by_sense(s)
sense_hyponyms[s] = nlp.read_hyponyms_by_sense(s)
sense_synonyms[s] = nlp.read_synonyms_by_sense(s)
sense_vectors[s] = nlp.get_pooling(s, sense_hypernyms,sense_synonyms,sense_hyponyms, word_vectors, VECTOR_DIM)
if ( word_vectors.has_key(w)):
sense_vectors[s] = sense_vectors[s] + word_vectors[w]
word_pool[w] = nlp.get_pooling(w, word_hypernyms, word_synonyms, word_hyponyms, word_vectors, VECTOR_DIM)
test_sense_vectors()
print "pre_cost",pre_cost, "cost", cost
if ( (pre_cost - cost) <= 1e-8 ):
s_vecs = pre_vecs
break
pre_vecs = s_vecs
i = 0
for s in senses[w]:
sense_vectors[s] = s_vecs[i]
i = i + 1
test_sense_vectors()
if __name__ == "__main__":
word_vectors = nlp.read_word_vectors(VECTOR_DIR + VECTOR_NAME)
print "LEARNING_RATE", L_RATE
word_pairs = nlp.read_csv(CSV_DIR + CSV_NAME)
vocab = []
for p in word_pairs:
vocab.append(p[0])
vocab.append(p[1])
vocab = list(set(vocab))
for w in vocab:
word_hypernyms[w] = nlp.read_hypernyms(w)
word_hyponyms[w] = nlp.read_hyponyms(w)
word_synonyms[w] = nlp.read_synonyms(w)
senses[w] = nlp.read_senses(w)
for s in senses[w]:
sense_hypernyms[s] = nlp.read_hypernyms_by_sense(s)
sense_hyponyms[s] = nlp.read_hyponyms_by_sense(s)
sense_synonyms[s] = nlp.read_synonyms_by_sense(s)
word_pool[w] = nlp.get_pooling(w, word_hypernyms, word_synonyms, word_hyponyms, word_vectors)
train_NN()
for p in word_pairs:
vocab.append(p[0].lower())
vocab.append(p[1].lower())
vocab = list(set(vocab))
for w in vocab:
word_hypernyms[w] = nlp.read_hypernyms(w)
word_hyponyms[w] = nlp.read_hyponyms(w)
word_synonyms[w] = nlp.read_synonyms(w)
senses[w] = nlp.read_senses(w)
for s in senses[w]:
sense_vectors[s] = np.zeros(VECTOR_DIM)
sense_hypernyms[s] = nlp.read_hypernyms_by_sense(s)
sense_hyponyms[s] = nlp.read_hyponyms_by_sense(s)
sense_synonyms[s] = nlp.read_synonyms_by_sense(s)
sense_pool[s] = nlp.get_pooling(s, sense_hypernyms, sense_synonyms,
sense_hyponyms, word_vectors,
VECTOR_DIM)
if (word_vectors.has_key(w)):
sense_pool[s] = sense_pool[s] + word_vectors[w]
sense_vectors[s] = sense_pool[s]
for l in s.lemmas():
word = str(l.name())
word_hypernyms[word] = nlp.read_hypernyms(word)
word_hyponyms[word] = nlp.read_hyponyms(word)
word_synonyms[word] = nlp.read_synonyms(word)
word_pool[word] = nlp.get_pooling(word, word_hypernyms,
word_synonyms, word_hyponyms,
word_vectors, VECTOR_DIM)
word_pool[w] = nlp.get_pooling(w, word_hypernyms, word_synonyms,
word_hyponyms, word_vectors, VECTOR_DIM)
s_vecs, s_star = calc_NN(w, p_w = para_w, p_u = para_u, p_v = para_v, p_b = para_b)
cost = cost_function(word_pool[w], s_star)
print "pre_cost",pre_cost, "cost", cost
if ( cost > pre_cost ):
s_vecs = pre_vecs
break
i = 0
for s in senses[w]:
sense_vectors[s] = s_vecs[i]
i = i + 1
if __name__ == "__main__":
word_vectors = nlp.read_word_vectors(VECTOR_DIR + VECTOR_NAME)
print "LEARNING_RATE", L_RATE
word_pairs = nlp.read_csv(CSV_DIR + CSV_NAME)
vocab = []
for p in word_pairs:
vocab.append(p[0])
vocab.append(p[1])
vocab = list(set(vocab))
for w in vocab:
word_hypernyms[w] = nlp.read_hypernyms(w)
word_hyponyms[w] = nlp.read_hyponyms(w)
word_synonyms[w] = nlp.read_synonyms(w)
senses[w] = nlp.read_senses(w)
for s in senses[w]:
sense_hypernyms[s] = nlp.read_hypernyms_by_sense(s)
sense_hyponyms[s] = nlp.read_hyponyms_by_sense(s)
sense_synonyms[s] = nlp.read_synonyms_by_sense(s)
word_pool[w] = nlp.get_pooling(w, word_hypernyms, word_synonyms, word_hyponyms, word_vectors)
train_NN()