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
|
"""
Spyder Editor
Author: Yihangbao
"""
from __future__ import print_function
import pandas as pd
import numpy as np
from keras.models import Sequential
from keras.optimizers import SGD,RMSprop,Adam
from keras.layers import Dense,Activation,Dropout
raw_train = pd.read_csv('train.csv', index_col=0)
raw_train['is_test'] = 0
raw_test = pd.read_csv('test.csv', index_col=0)
raw_test['is_test'] = 1
all_data = pd.concat((raw_train,raw_test), axis=0)
def get_titles(df):
df['Title'] = df['Name'].map(lambda name:name.split(',')[1].split('.')[0].strip())
Title_Dictionary = {
"Capt": "Officer",
"Col": "Officer",
"Major": "Officer",
"Jonkheer": "Royalty",
"Don": "Royalty",
"Sir" : "Royalty",
"Dr": "Officer",
"Rev": "Officer",
"the Countess":"Royalty",
"Dona": "Royalty",
"Mme": "Mrs",
"Mlle": "Miss",
"Ms": "Mrs",
"Mr" : "Mr",
"Mrs" : "Mrs",
"Miss" : "Miss",
"Master" : "Master",
"Lady" : "Royalty"
}
df['Title'] = df.Title.map(Title_Dictionary)
df = df.drop(['Name'], axis=1)
return df
def get_dummy_cats(df):
return(pd.get_dummies(df,columns=['Title','Pclass','Sex','Embarked','Cabin','Cabin_letter']))
def get_cabin_letter(df):
df['Cabin'].fillna('Z', inplace=True)
df['Cabin_letter'] = df['Cabin'].str[0]
return(df)
def process_data(df):
df = get_titles(df)
df['Embarked'].fillna('S', inplace=True)
df = get_cabin_letter(df)
df = df.drop(['Ticket','Fare'], axis=1)
df = get_dummy_cats(df)
return df
proc_data = process_data(all_data)
proc_train = proc_data[proc_data['is_test'] == 0]
proc_test = proc_data[proc_data['is_test'] == 1]
for_age_train = proc_data.drop(['Survived','is_test'], axis=1).dropna(axis=0)
X_train_age = for_age_train.drop('Age',axis=1)
Y_train_age = for_age_train['Age']
tmodel = Sequential()
tmodel.add(Dense(input_dim=X_train_age.shape[1], units=128, kernel_initializer='normal', bias_initializer='zeros'))
tmodel.add(Activation('relu'))
for i in range(0,8):
tmodel.add(Dense(units=64, kernel_initializer='normal',bias_initializer='zeros'))
tmodel.add(Dropout(.25))
tmodel.add(Dense(units=1))
tmodel.add(Activation('linear'))
tmodel.compile(loss='mean_squared_error', optimizer='rmsprop')
tmodel.fit(X_train_age.values, Y_train_age.values,epochs=600,verbose=2)
train_data = proc_train
train_data.loc[train_data['Age'].isnull()]
to_pred = train_data.loc[train_data['Age'].isnull()].drop(['Age','Survived','is_test'], axis=1)
p = tmodel.predict(to_pred.values)
train_data.loc[train_data['Age'].isnull(),'Age'] = p
test_data = proc_test
to_pred = test_data.loc[test_data['Age'].isnull()].drop(['Age','Survived','is_test'], axis=1)
p = tmodel.predict(to_pred.values)
test_data.loc[test_data['Age'].isnull(),'Age'] = p
y = pd.get_dummies(train_data['Survived'])
x= train_data.drop(['Survived','is_test'],axis=1)
model = Sequential()
model.add(Dense(input_dim=x.shape[1], units=128, kernel_initializer='normal', bias_initializer='zeros'))
model.add(Activation('relu'))
for i in range(15):
model.add(Dense(units=128, kernel_initializer='normal' ,bias_initializer='zeros'))
model.add(Activation('relu'))
model.add(Dropout(.40))
model.add(Dense(units=2))
model.add(Activation('softmax'))
model.compile(loss='categorical_crossentropy',optimizer='adam',metrics=['accuracy'])
model.fit(x.values,y.values,epochs=500, verbose=2)
p_survived = model.predict_classes(test_data.drop(['Survived','is_test'],axis=1).values)
submission = pd.DataFrame()
submission['PassengerID'] = test_data.index
submission['Survived'] = p_survived
submission.to_csv('ans1.csv',index=False)
|
