【Keras学习笔记】6：MLP预测Titanic数据集

读取数据和预处理

import keras
from keras import layers
from matplotlib import pyplot as plt
import numpy as np
import pandas as pd
%matplotlib inline

Using TensorFlow backend.

data = pd.read_csv("./data/tt_train.csv")
x = data[['Survived', 'Pclass', 'Sex', 'Age', 'SibSp',
       'Parch', 'Fare', 'Embarked']]
x = x.copy()
# 给Embarked添加one-hot编码
x.loc[:,'Embarked_S']=(x.Embarked=='S').astype('int')
x.loc[:,'Embarked_C']=(x.Embarked=='C').astype('int')
x.loc[:,'Embarked_Q']=(x.Embarked=='Q').astype('int')
del x['Embarked']
# 给Sex添加one-hot编码
x = x.join(pd.get_dummies(x.Sex))
del x['Sex']
# 给Age的缺失值填充均值
x['Age'] = x.Age.fillna(x.Age.mean())
# 给Pclass添加one-hot编码
x.loc[:,'P1'] = (x.Pclass==1).astype('int') 
x.loc[:,'P2'] = (x.Pclass==2).astype('int') 
x.loc[:,'P3'] = (x.Pclass==3).astype('int') 
del x['Pclass']
# 现在预处理完成了,把预测值取出来,并在x中把它删掉
y = data.Survived
del x['Survived']
x.shape, y.shape

((891, 12), (891,))

建立和训练模型

model = keras.Sequential()
# MLP是多层的
model.add(layers.Dense(32, input_dim=12, activation='relu'))
model.add(layers.Dense(32, activation='relu')) # 隐含层输入维度会自动和上一层一样
model.add(layers.Dense(1, activation='sigmoid')) # 因为是二分类的,所以只要输出1维然后用sigmoid激活

WARNING:tensorflow:From E:\MyProgram\Anaconda\envs\krs\lib\site-packages\tensorflow\python\framework\op_def_library.py:263: colocate_with (from tensorflow.python.framework.ops) is deprecated and will be removed in a future version.
Instructions for updating:
Colocations handled automatically by placer.

model.summary()

_________________________________________________________________
Layer (type)                 Output Shape              Param #   
=================================================================
dense_1 (Dense)              (None, 32)                416       
_________________________________________________________________
dense_2 (Dense)              (None, 32)                1056      
_________________________________________________________________
dense_3 (Dense)              (None, 1)                 33        
=================================================================
Total params: 1,505
Trainable params: 1,505
Non-trainable params: 0
_________________________________________________________________

model.compile(
    optimizer='adam',
    loss='binary_crossentropy',
    metrics=['acc']
)

history = model.fit(x, y, epochs=300, verbose=0)

WARNING:tensorflow:From E:\MyProgram\Anaconda\envs\krs\lib\site-packages\tensorflow\python\ops\math_ops.py:3066: to_int32 (from tensorflow.python.ops.math_ops) is deprecated and will be removed in a future version.
Instructions for updating:
Use tf.cast instead.

绘制训练过程

plt.plot(range(300),history.history.get('loss'))

[<matplotlib.lines.Line2D at 0x1622de48>]

plt.plot(range(300),history.history.get('acc'))

[<matplotlib.lines.Line2D at 0x162bd9b0>]

导出预测值以提交到Kaggle

# 读入数据,做相同的预处理
data = pd.read_csv("./data/tt_test.csv")
xt = data[['Pclass', 'Sex', 'Age', 'SibSp',
       'Parch', 'Fare', 'Embarked']]
xt = xt.copy()
# 给Embarked添加one-hot编码
xt.loc[:,'Embarked_S']=(xt.Embarked=='S').astype('int')
xt.loc[:,'Embarked_C']=(xt.Embarked=='C').astype('int')
xt.loc[:,'Embarked_Q']=(xt.Embarked=='Q').astype('int')
del xt['Embarked']
# 给Sex添加one-hot编码
xt = xt.join(pd.get_dummies(xt.Sex))
del xt['Sex']
# 给Age的缺失值填充均值
x['Age'] = xt.Age.fillna(xt.Age.mean())
# 给Pclass添加one-hot编码
xt.loc[:,'P1'] = (xt.Pclass==1).astype('int') 
xt.loc[:,'P2'] = (xt.Pclass==2).astype('int') 
xt.loc[:,'P3'] = (xt.Pclass==3).astype('int') 
del xt['Pclass']
xt.shape

(418, 12)

# 计算预测值
predictions = model.predict(xt)

# 生成提交csv
submission = pd.DataFrame({"PassengerId": data["PassengerId"], "Survived": (predictions.flatten()>0.5).astype('int')})
submission.to_csv("./data/tt_upload.csv", index=False)

E:\MyProgram\Anaconda\envs\krs\lib\site-packages\ipykernel_launcher.py:2: RuntimeWarning: invalid value encountered in greater