-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathhahaha.py
129 lines (104 loc) · 5.07 KB
/
hahaha.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
import pandas as pd
import numpy as np
import time
from A_3_feature_engine import *
from D_depend import *
from joblib import dump, load
# coding=utf-8
import pandas as pd
import numpy as np
from D_depend import *
from joblib import dump, load
from sklearn.model_selection import StratifiedShuffleSplit
from sklearn.metrics import log_loss
from sklearn.preprocessing import LabelEncoder
from sklearn.ensemble import RandomForestClassifier
from datetime import datetime
import time
import os
raw_train = load('./dump/train.joblib_dat')
raw_test = load('./dump/test.joblib_dat')
# process cate camgaignID for train and test !!!
# col_name = ['camgaignID', 'advertiserID', 'creativeID', 'adID']
# clilc_rate = [
# [0, 0.05, 0.10, 0.15, 0.20, 0.25, 0.30, 0.35, 0.4],
# [0, 0.03, 0.11],
# [0, 0.05, 0.07, 0.09, 0.10, 0.12, 0.14, 0.16, 0.18, 0.20, 0.25, 0.30, 0.35, 0.4],
# [0, 0.05, 0.07, 0.09, 0.10, 0.12, 0.14, 0.16, 0.18, 0.20, 0.25, 0.30, 0.35, 0.4]
# ]
# for i in range(len(col_name)):
# print("--------------->分类" + col_name[i])
# raw_train, raw_test = cate_(raw_train, raw_test, None, None, col_name[i], clilc_rate[i])
# process train set
print("--------------->统计train is_app_actions")
app_actions = pd.read_csv('./data/user_app_actions.csv')
app_actions['is_app_actions'] = 1
raw_train = pd.merge(raw_train, app_actions, on=['userID', 'appID'], sort=False, how='left')
raw_train['is_app_actions'] = raw_train['is_app_actions'].fillna(0)
raw_train['is_app_actions'] = raw_train['is_app_actions'].astype('int')
raw_train['town_sheng'] = (raw_train.hometown / 100).astype(int)
raw_train['residence_sheng'] = (raw_train.residence / 100).astype(int)
raw_train['con_tele'] = (raw_train.connectionType * 10 + raw_train.telecomsOperator).astype(int)
raw_train['age_edu'] = (raw_train.education * 100 + raw_train.age).astype(int)
raw_train['hour'] = (raw_train.clickTime.values / 100).astype(int) % 100
# same for test set
print("--------------->统计test is_app_actions")
raw_test = pd.merge(raw_test, app_actions, on=['userID', 'appID'], sort=False, how='left')
raw_test['is_app_actions'] = raw_test['is_app_actions'].fillna(0)
raw_test['is_app_actions'] = raw_test['is_app_actions'].astype('int')
raw_test['town_sheng'] = (raw_test.hometown / 100).astype(int)
raw_test['residence_sheng'] = (raw_test.residence / 100).astype(int)
raw_test['con_tele'] = (raw_test.connectionType * 10 + raw_test.telecomsOperator).astype(int)
raw_test['age_edu'] = (raw_test.education * 100 + raw_test.age).astype(int)
raw_test['hour'] = (raw_test.clickTime.values / 100).astype(int) % 100
# col = ['connectionType','telecomsOperator','camgaignID',
# 'appID', 'appPlatform', 'is_app_actions','advertiserID',
# 'age', 'gender','education', 'marriageStatus', 'haveBaby', 'hometown',
# 'residence','positionID', 'sitesetID', 'positionType']
col = ['connectionType','telecomsOperator','camgaignID',
'appID', 'appPlatform', 'is_app_actions','advertiserID',
'age', 'gender','education', 'marriageStatus', 'haveBaby','hometown',
'residence','positionID', 'sitesetID', 'positionType','hour','advertiserID',
'creativeID', 'adID']
raw_train = raw_train[col]
raw_test = raw_test[col]
# process the new value in test
for i in raw_train.columns.values:
raw_test[(raw_test['hour'].isin(raw_train['hour'].values) == False).values] = 0
# dump(raw_train, "./dump/new_train.joblib_dat")
# dump(raw_test, "./dump/new_test.joblib_dat")
train = pd.read_csv('./data/train.csv')
test = pd.read_csv('./data/test.csv')
trainY = train.label.values
# feature_train = load("./dump/new_train.joblib_dat")
# feature_test = load('./dump/new_test.joblib_dat')
feature_train = raw_train
feature_test = raw_test
# --------------------------去掉第30天------------------------
feature_train = feature_train[train.clickTime.values < 300000]
trainY = trainY[train.clickTime.values < 300000]
print('# ------------------Cross validation..-----------------------')
max_depth = 14
n_splits = 3
test_size = 0.1
train_size = 0.2
# model = RandomForestClassifier(max_depth=max_depth, n_jobs=-1, random_state=777)
model = RandomForestClassifier(n_estimators=32, max_depth=14, min_samples_split=100, min_samples_leaf=10, random_state=0,
criterion='entropy', max_features=8, verbose = 1, n_jobs=-1, bootstrap=False)
trn_scores = []
vld_scores = []
sss = StratifiedShuffleSplit(n_splits=n_splits, test_size=test_size, train_size=train_size, random_state=777)
for i, (t_ind, v_ind) in enumerate(sss.split(feature_train, trainY)):
print('# Iter {} / {}'.format(i + 1, n_splits))
x_trn = feature_train.values[t_ind]
y_trn = trainY[t_ind]
x_vld = feature_train.values[v_ind]
y_vld = trainY[v_ind]
model.fit(x_trn, y_trn)
score = log_loss(y_trn, model.predict_proba(x_trn))
trn_scores.append(score)
score = log_loss(y_vld, model.predict_proba(x_vld))
vld_scores.append(score)
print("max_depth: %d n_splits: %d test_size: %f" % (max_depth, n_splits, test_size))
print('# TRN logloss: {}'.format(np.mean(trn_scores)))
print('# VLD logloss: {}'.format(np.mean(vld_scores)))