def main():
# filename = 'final_test.csv'
pre()
clf = joblib.load("model.pkl")
df = pd.read_csv('testing.csv')
df.drop(['Unnamed: 0'], axis=1, inplace=True)
df = df.dropna()
columnsTitles = ['fit', 'body_type', 'category', 'weight', 'rating', 'height',
'age', 'bust_size', 'cup_size_start_in_cms', 'cup_size_end_in_cms']
df = df.reindex(columns=columnsTitles)
test = np.array(df)
Y_pred = clf.predict(test)
print(Y_pred)
size = {0: 'XS', 1: 'S', 2: 'M', 3: 'L', 4: 'XL', 5: 'XXL'}
x = np.vectorize(size.get)(Y_pred)
return x
plt.rc('xtick', labelsize=5)
plt.autoscale(enable=True, axis='x', tight=True)
plt.tight_layout()
plt.show()
# Plotting training data-rainfall
plt.plot(train_date, train_rain)
plt.title("Rainfall")
plt.xticks(rotation=90)
plt.rc('xtick', labelsize=5)
plt.autoscale(enable=True, axis='x', tight=True)
plt.tight_layout()
plt.show()
# Preprocessing Data
pre(train_temp)
pre(train_rain)
# Fitting Univariate Data-temperature
model_arima(train_temp, test_temp)
model_arma(train_temp, test_temp)
model_sarima(train_temp,test_temp)
# Fitting Univariate Data-rainfall
model_arima(train_rain, test_rain)
model_arma(train_rain, test_rain)
model_sarima(train_rain,test_rain)
# Fitting Multivariate Data
model_sarimax(train_temp,test_temp,train_rain,test_rain)
model_var(train_temp,test_temp,train_rain,test_rain)
for item in datetime:
a, b = item.split(' ')
date.append(a)
time.append(b)
# Data Pre-processing - Fill Missing Values
humidity = humidity.fillna(method='ffill', axis=0)
pressure = pressure.fillna(method='ffill', axis=0)
temp = temp.fillna(method='ffill', axis=0)
# Typecasting and Data Splitting
humidity = np.asarray(humidity)
train_humidity, test_humidity = humidity[0:36000:168], humidity[36000::168]
train_date, test_date = date[0:36000:168], date[36000::168]
print(np.shape(test_humidity), len(train_date))
# Plotting training data
plt.plot(train_date, train_humidity)
plt.title("Humidity")
plt.xticks(rotation=90)
plt.rc('xtick', labelsize=5)
plt.autoscale(enable=True, axis='x', tight=True)
plt.tight_layout()
plt.show()
# Preprocessing Data
pre(train_humidity)
# Fitting Data
model_arima(train_humidity, test_humidity)
])
pred = lgb_model.predict_proba(test[col])[:, 1]
test['predicted_score'] = pred
#processHasTrade(test,"predicted_score")
sub1 = test[['instance_id', 'predicted_score']]
sub = pd.read_csv("te.csv")
sub = pd.merge(sub, sub1, on=['instance_id'], how='left')
sub = sub.fillna(0)
sub[['instance_id', 'predicted_score']].to_csv('result0326.txt',
sep=" ",
index=False)
if __name__ == "__main__":
pre()
online = True
train = pd.read_csv("tr.csv")
test = pd.read_csv("te.csv")
data = pd.concat([train, test])
data = data.drop_duplicates(subset='instance_id')
data = base_process(data)
#data = process_prop(data,"item_property_list")
#svdSpareMat(data,"item_property_list")
#data = user_click_comm_level(data) #No obvious change
#data = comm_extra(data)- #cause bigger loss
#data = comm_gender(data)
data = brand_extra(data) #-Good feature
data = user_extra(data) #-Good feature
data = shop_extra(data) #-Good feature
#data = user_click_comm_level(data)
# -*- coding: utf8 -*-
import sys
import re
from preprocess import pre
reload(sys)
sys.setdefaultencoding("utf8")
if __name__ == "__main__":
test_str = u'二、变更上海市浦东新区人民法院（2013）浦民一（民）初字第30748号民事判决第十三项为：上述第四、七、八、九、十项，于本判决生效之日起十日内，由傅甲支付阎丙37，655.50元、姚丁37，655.50元、阎乙96，919.90元，合计172，230.90元（减去已支付的113，500元，实际应支付五万八千七百元九角）。'
print test_str
print pre(test_str)
def test_sep_senfin_pun(self):
self.assertEqual(pre("separate comma, and other.", "e"),
"separate comma , and other .")
def test_double_quot(self):
self.assertEqual(pre("he said ''i dont know'' ", "e"),
"he said '' i dont know ''")
def test_dash_in_bracket(self):
self.assertEqual(pre("(i-don't-know)", "e"), "( i - don't - know )")
def test_bracket(self):
self.assertEqual(pre("(i don't know)", "e"), "( i don't know )")
def test_leading_c_a(self):
self.assertEqual(pre("je t'aime", "f"), "je t' aime")
def test_leading_l_punctuation(self):
self.assertEqual(pre("l'election", "f"), "l' election")
