-
Notifications
You must be signed in to change notification settings - Fork 4
/
Copy pathlgb.py
181 lines (152 loc) · 6.02 KB
/
lgb.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
import pandas as pd
import numpy as np
from sklearn.preprocessing import MinMaxScaler, LabelEncoder
from sklearn.model_selection import train_test_split
from sklearn.feature_selection import VarianceThreshold
import lightgbm as lgb
from lightgbm import plot_importance
import matplotlib.pyplot as plt
import seaborn as sns
import os
import sklearn
import time
import random
import operator
from sklearn.model_selection import StratifiedKFold
from pandas import DataFrame as DF
start_time =time.time()
time_date = time.strftime('%Y-%m-%d', time.localtime(time.time()))
random.seed(1000)
'''
单模型+排名前80特征选择
'''
def my_score(estimator, X, y):
predicted = estimator.predict(X)
return 1/(1+np.sqrt(np.mean((predicted - y) ** 2)))
def get_pic(model, feature_name):
ans = DF()
ans['name'] = feature_name
ans['score'] = model.feature_importances_
# print(ans[ans['score']>0].shape)
return ans.sort_values(by=['score'], ascending=False).reset_index(drop=True)
def gey_100_impot_feature(train,y,nums=80):
print('nums:',nums)
#
lgb_model = lgb.LGBMRegressor(n_estimators=120)
# cv_model = cv(lgb_model, train_data[feature_name], train_label, cv=10, scoring='f1')
lgb_model.fit(train, y)
feature_name1 = train.columns
#print('用到的特征：：：',feature_name.ix[:nums])
return list(set(get_pic(lgb_model,feature_name1).head(nums)['name']))
def UseLightGBM():
'''
:return:使用单模型做5折cv
'''
lgbm_submission = pd.read_csv('./data/base_data/submit_example.csv',header=None)
lgbm_submission.columns=['A','B']
print('###:',)
#train_flg = pd.read_csv(os.getcwd() + '/data/train/train_flg.csv', sep='\t')
train = pd.read_csv( './data/do_feat_data_3/train.csv')
test = pd.read_csv( './data/do_feat_data_3/test.csv')
#tuihuo_train = pd.read_csv( './data/do_feat_data/train_tuihuo.csv')
y = train.pop('发电量')
imp_feature_nam=gey_100_impot_feature(train,y)
print('imp_feature_nam',imp_feature_nam)
#print(':::::::',tuihuo_train.columns)
test=test[imp_feature_nam]
#id是有用的
#train.pop('ID')
#test.pop('ID')
print(train.columns)
# train=np.log(train)
# test=np.log(test)
# print('')
#print('train:',train['发电量'])
train=train[imp_feature_nam]
print('train:',train.head())
print('test:',test.head())
#################################### 交叉检验 #####################################
N = 10
xx_cv = []
xx_pre = []
xx_beat = {}
kf = sklearn.model_selection.KFold(n_splits=5, shuffle=True, random_state=42)
modelScores = []
for train_index, test_index in kf.split(train):
X_train, X_test, y_train, y_test = train.values[train_index], train.values[test_index], y[train_index], y[test_index]
# create dataset for lightgbm
lgb_train = lgb.Dataset(X_train, y_train)
lgb_eval = lgb.Dataset(X_test, y_test, reference=lgb_train)
# specify your configurations as a dict
params = {
'task': 'train',
'boosting_type': 'gbdt',
'objective': 'regression_l2',
'metric': {'l2', 'l1'},
'min_data_in_leaf': 20,
'feature_fraction': 1,
'bagging_freq': 1,
'verbose': 1,
'seed':50
}
print('Start training...')
# train
gbm = lgb.train(params,
lgb_train,
num_boost_round=40000,
valid_sets=lgb_eval,
early_stopping_rounds=50,
verbose_eval=250
)
#y_pred = gbm.predict(X_test, num_iteration=gbm.best_iteration)
#将使用这个最好的得分作为xx_cv下的附加得分
xx_cv.append(my_score(gbm,X_test,y_test))
print('得分是：',xx_cv)
xx_pre.append(gbm.predict(test))
#图像展示lgb模型训练时候的特征重要度,并保存起来(两种保存方式:原序和降序)
# plot_importance(gbm)
# plt.show()
FeatureScore=gbm.feature_importance(importance_type='split')
df=pd.DataFrame({'Feature':train.columns.tolist(),'importance':FeatureScore.tolist()})
df.to_csv(os.getcwd() +'/record/lgbm_feautre_important_{0}.csv'.format(str(time_date)), index=False)
sort_df=df.sort_values(by='importance',ascending=False)
sort_df.to_csv(os.getcwd() +'/record/sort_lgbm_feautre_important_{0}.csv'.format(str(time_date)), index=False)
# df.plot(kind='bar',title='feature important')
# plt.show()
#获取最好cv得分的序号，使用这个序号去找到序号下对应的预测结果
sorted_cv = sorted(enumerate(xx_cv), key=lambda x: x[1], reverse=True)
best_cv_index=sorted_cv[0][0]
print('平均得分是：', np.mean(xx_cv))
#################################### 直接生成概率结果 #####################################
params = {
'task': 'train',
'boosting_type': 'gbdt',
'objective': 'regression_l2',
'metric': {'l2', 'l1'},
'min_data_in_leaf': 20,
'feature_fraction': 1,
'bagging_freq': 1,
'verbose': 0,
'seed': 50
}
print('Start training...')
# train
lgb_train = lgb.Dataset(train, y)
gbm = lgb.train(params,
lgb_train,
num_boost_round=500)
pred_y=gbm.predict(test)
res = pd.DataFrame()
res['A'] = list(lgbm_submission['A'])
print('!!!!!!',len(res['A']))
print('!222!!!!!:', len(xx_pre[best_cv_index]))
res['B'] = list(pred_y)
#print(res[res['RST'] == 1].shape)
#################################### 提交和线下结果展示 #####################################
res.to_csv(os.getcwd() +'/submit/%s_%s.csv'% (str(time_date), str(np.mean(xx_cv)).split('.')[1]),
index=False, header=None)
print('end tiem', time.time() - start_time)
print('info')
print('线下成绩约', np.mean(xx_cv))
if __name__=='__main__':
UseLightGBM()