#学習済み
#http://cedro3.com/ai/word2vec-gensim/
#http://aial.shiroyagi.co.jp/2017/02/japanese-word2vec-model-builder/
#http://www.cl.ecei.tohoku.ac.jp/~m-suzuki/jawiki_vector/

 
import numpy as np
import matplotlib as mpl
import matplotlib.pyplot as plt


# モデルを読み込む

import codecs
from gensim.models import word2vec
model = word2vec.Word2Vec.load("ptenken_jpn.model")

# 亀裂と類似している単語を見る
similar_words = model.wv.most_similar(positive=["亀裂"], topn=30)
#similar_words = model.wv.most_similar(positive=["crack"], topn=30)
#similar_words = model.wv.most_similar(positive=["応力","亀裂"], topn=30)
#similar_words = model.wv.most_similar(positive=["亀裂"],negative=["ひび割れ"], topn=30)
print(*[" ".join([v, str("{:.2f}".format(s))]) for v, s in similar_words], sep="\n")


'''
#乾モデルを使う場合
from gensim.models import KeyedVectors

model_dir = './entity_vector.model.bin'
model = KeyedVectors.load_word2vec_format(model_dir, binary=True)

#results = model.most_similar(u'[積雪]')
#results = model.most_similar(positive=[u'[道路]',u'[積雪]'])
#results = model.most_similar(positive=[u'[渋滞]',u'[原因]'])
#results = model.most_similar(positive=[u'[路面]',u'[雪]'],negative=[u'[事故]'])
results = model.most_similar(positive=[u'[世界]'],negative=[u'[知性]'])


#白ヤギモデル
from gensim.models.word2vec import Word2Vec

model_path = 'word2vec.gensim.model'
model = Word2Vec.load(model_path)

results = model.most_similar(u'[雪]')

FONT="ipaexg.ttf"

'''