import pandas
import sys
from chchao.like_test import like_test
profile = pandas.read_csv('data/TCSS555/Train/Profile/Profile.csv')
relation = pandas.read_csv('data/TCSS555/Train/Relation/Relation.csv')
a = []
b = []
_a = []
_b = []
n_size = 9500
n_train = 8500
knn = 5
x = like_test()
for i in range(0, n_size):
l = x.lr_formating(relation, profile['userid'][i])
a.append(l)
b.append(x.lr_age_format(profile['age'][i]))
if 1 in a[i]:
_a.append(a[i])
_b.append(b[i])
sys.stdout.write("training : %4d/%04d\r"%(i,n_size))
sys.stdout.flush()
print("knn : %d"%knn)
print("n_train : %d"%n_train)
print("len(_a):%d len(_b):%d"%(len(_a), len(_b)))
# from sklearn.decomposition import TruncatedSVD
try:
relation = pandas.read_csv(input_dir+'relation/relation.csv')
print("reading relation.csv successed.")
except:
print("Error: reading relation.csv failed.")
exit()
try:
oxford_csv = pandas.read_csv(input_dir+'oxford.csv')
print("reading oxford.csv successed.")
except:
print("Warning: reading oxford.csv failed.")
baseline = baseline()
like_test = like_test()
cf = open('/home/itadmin/MLProject/clf_age_SGDC_liketext.pickle', 'rb')
age_predict = pickle.load(cf)
# cf = open('/home/itadmin/MLProject/clf_like_mnb_gender.pickle', 'rb')
# g_predict = pickle.load(cf)
g_predict = gender()
len_profile = len(profile)
try:
ope_p = text_ope_knn_test()
except:
