-
Notifications
You must be signed in to change notification settings - Fork 1
/
RFtrain.py
134 lines (124 loc) · 4.37 KB
/
RFtrain.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
import csv
import numpy as np
import string
from sklearn.ensemble import RandomForestClassifier
from sklearn.naive_bayes import GaussianNB
from sklearn.metrics import roc_curve, auc
import pandas as p
import matplotlib.pyplot as plt
def outlier(data,col,m=2):
st = np.std(data[:,col])
me = np.mean(data[:,col])
dele = []
for i in xrange(len(data)):
if not (abs(data[i][col] - me) < m * st):
dele.extend([i])
return np.delete(data,dele,0)
def main():
train = p.read_table('train.tsv').replace('?',0)
# target = np.array(train)[:,-1]
train['alchemy_category'] = train.groupby('alchemy_category').grouper.group_info[0]
train['alchemy_category_score'] = train['alchemy_category_score'].astype(float)
# train = np.array(train)[:,:-1]
train = np.array(train)[:,3:]
test = p.read_table('test.tsv').replace('?',0)
test['alchemy_category'] = test.groupby('alchemy_category').grouper.group_info[0]
test['alchemy_category_score'] = test['alchemy_category_score'].astype(float)
valid_index = list(np.array(test)[:,1])
test = np.array(test)[:,3:]
for i in range(2,23):
if i == 9:
continue
try:
test = train
test = outlier(test,i)
target = test[:,-1]
test = test[:,:-1]
print len(test)
# alchemy_category_set = {}
# #read train data
# train = []
# target = []
# with open("train.tsv", 'rb') as csvfile:
# reader = csv.reader(csvfile, dialect='excel-tab')
# reader.next() #skip the header
# for row in reader:
# line = row[3:len(row)-1]
# train.append(line)
# if row[len(row)-1] == '?':
# target.append(0)
# else:
# target.append(int(row[len(row)-1]))
# if row[3] not in alchemy_category_set:
# alchemy_category_set[row[3]] = len(alchemy_category_set)
# #read valid data
# valid = []
# valid_index = []
# with open("test.tsv", 'rb') as csvfile:
# reader = csv.reader(csvfile, dialect='excel-tab')
# reader.next() #skip the header
# for row in reader:
# line = row[3:len(row)]
# valid.append(line)
# valid_index.append(row[1])
# if row[3] not in alchemy_category_set:
# alchemy_category_set[row[3]] = len(alchemy_category_set)
# #expand the alchemy_category
# for idx in range(len(train)):
# line = train[idx]
# alchemy_category = [0 for i in range(len(alchemy_category_set))]
# alchemy_category_idx = alchemy_category_set[line[0]]
# alchemy_category[alchemy_category_idx] = 1
# del line[0]
# line = [string.atof(x) if x != '?' else 0 for x in line]
# line = line + alchemy_category
# train[idx] = line
# for idx in range(len(valid)):
# line = valid[idx]
# alchemy_category = [0 for i in range(len(alchemy_category_set))]
# alchemy_category_idx = alchemy_category_set[line[0]]
# alchemy_category[alchemy_category_idx] = 1
# del line[0]
# line = [string.atof(x) if x != '?' else 0 for x in line]
# line = line + alchemy_category
# valid[idx] = line
r = []
r.append([0,0.000])
# for j in range(9,10):
n = int((8.5*len(train))/10)
X_train = test[:n]
X_test = test[n:]
y_train = target[:n]
y_test = target[n:]
# run the model
classifier = RandomForestClassifier(n_estimators=1000,verbose=0,n_jobs=20,min_samples_split=5,random_state=1034324)
# classifier = GaussianNB()
classifier.fit(X_train, y_train)
pred = classifier.predict_proba(X_test)
fpr, tpr, thresholds = roc_curve(y_test,pred[:,1])
roc_auc = auc(fpr, tpr)
print("%d Area under the ROC curve : %f" %(i,roc_auc))
except TypeError, ValueError:
continue
# r.append([j,roc_auc])
# plt.grid(True)
# print r
# x = [i[0]*10 for i in r]
# y = [i[1]*100 for i in r]
# plt.plot(x,y,linewidth=3)
# plt.axis([0,100,0,100])
# plt.xlabel("training % data")
# plt.ylabel('Accuracy (CV score k=20)')
# plt.show()
# gnb.fit(X_train, y_train)
# pred = gnb.predict(X_test)
# fpr, tpr, thresholds = roc_curve(y_test,pred)
# roc_auc = auc(fpr, tpr)
# print("Area under the ROC curve : %f" % roc_auc)
# write
writer = csv.writer(open("predictions", "w"), lineterminator="\n")
rows = [x for x in zip(valid_index, classifier.predict(test))]
writer.writerow(("urlid","label"))
writer.writerows(rows)
if __name__=="__main__":
main()