Keras的泰坦尼克号的生存率的数据分析

 

# coding: utf-8

# In[1]:


import urllib.request
import os


# In[2]:


url="http://biostat.mc.vanderbilt.edu/wiki/pub/Main/DataSets/titanic3.xls"
filepath="titanic3.xls"
if not os.path.isfile(filepath):
    result=urllib.request.urlretrieve(url,filepath)
    print('downloaded:',result)


# In[3]:


import numpy
import pandas as pd


# In[4]:


all_df = pd.read_excel(filepath)


# In[5]:


all_df[:5]


# In[6]:


cols=['survived','name','pclass' ,'sex', 'age', 'sibsp',
      'parch', 'fare', 'embarked']
all_df=all_df[cols]
all_df[:5]


# In[7]:


all_df.isnull().sum()


# In[8]:


df=all_df.drop(['name'], axis=1)
df[:20]


# In[9]:


age_mean = df['age'].mean()
df['age'] = df['age'].fillna(age_mean)
df[:20]


# In[10]:


fare_mean = df['fare'].mean()
df['fare'] = df['fare'].fillna(fare_mean)


# In[11]:


df['sex']= df['sex'].map({'female':0, 'male': 1}).astype(int)


# In[12]:


df[:2]


# In[13]:


x_OneHot_df = pd.get_dummies(data=df,columns=["embarked" ])


# In[14]:


x_OneHot_df[:2]


# In[15]:


ndarray = x_OneHot_df.values
ndarray.shape


# In[16]:


ndarray[:2]


# In[17]:


Label = ndarray[:,0]
Features = ndarray[:,1:]


# In[18]:


Features[:2]


# In[19]:


from sklearn import preprocessing
minmax_scale = preprocessing.MinMaxScaler(feature_range=(0, 1))
scaledFeatures=minmax_scale.fit_transform(Features)
scaledFeatures[:2]


# In[20]:


msk = numpy.random.rand(len(all_df)) < 0.8
train_df = all_df[msk]
test_df = all_df[~msk]


# In[21]:


msk


# In[22]:


print('total:',len(all_df),
      'train:',len(train_df),
      'test:',len(test_df))


# In[23]:


def PreprocessData(raw_df):
    df=raw_df.drop(['name'], axis=1)
    age_mean = df['age'].mean()
    df['age'] = df['age'].fillna(age_mean)
    fare_mean = df['fare'].mean()
    df['fare'] = df['fare'].fillna(fare_mean)
    df['sex']= df['sex'].map({'female':0, 'male': 1}).astype(int)
    x_OneHot_df = pd.get_dummies(data=df,columns=["embarked" ])

    ndarray = x_OneHot_df.values
    Features = ndarray[:,1:]
    Label = ndarray[:,0]

    minmax_scale = preprocessing.MinMaxScaler(feature_range=(0, 1))
    scaledFeatures=minmax_scale.fit_transform(Features)    
    
    return scaledFeatures,Label


# In[24]:


train_Features,train_Label=PreprocessData(train_df)
test_Features,test_Label=PreprocessData(test_df)


# In[25]:


train_Features[:2]


# In[26]:


train_Label[:2]


# In[27]:


from keras.models import Sequential
from keras.layers import Dense,Dropout


# In[28]:


model = Sequential()
model.add(Dense(units=40, input_dim=9, 
                kernel_initializer='uniform', 
                activation='relu'))
model.add(Dense(units=30, 
                kernel_initializer='uniform', 
                activation='relu'))
model.add(Dense(units=1, 
                kernel_initializer='uniform',
                activation='sigmoid'))
model.summary()


# In[29]:


model.compile(loss='binary_crossentropy', 
              optimizer='adam', metrics=['accuracy'])
train_history =model.fit(x=train_Features, 
                         y=train_Label, 
                         validation_split=0.1, 
                         epochs=30, 
                         batch_size=30,verbose=2)


# In[30]:


import matplotlib.pyplot as plt
def show_train_history(train_history,train,validation):
    plt.plot(train_history.history[train])
    plt.plot(train_history.history[validation])
    plt.title('Train History')
    plt.ylabel(train)
    plt.xlabel('Epoch')
    plt.legend(['train', 'validation'], loc='upper left')
    plt.show()
show_train_history(train_history,'acc','val_acc')
show_train_history(train_history,'loss','val_loss')


# In[31]:


scores = model.evaluate(x=test_Features, 
                        y=test_Label)
scores


# In[32]:


Jack = pd.Series([0 ,'Jack',3, 'male'  , 23, 1, 0,  5.0000,'S'])
Rose = pd.Series([1 ,'Rose',1, 'female', 20, 1, 0, 100.0000,'S'])
JR_df = pd.DataFrame([list(Jack),list(Rose)],  
                  columns=['survived', 'name','pclass', 'sex', 
                   'age', 'sibsp','parch', 'fare','embarked'])
all_df=pd.concat([all_df,JR_df])
all_df[-2:]


# In[33]:


all_Features,Label=PreprocessData(all_df)
all_probability=model.predict(all_Features)
all_probability[:10]


# In[34]:


pd=all_df
pd.insert(len(all_df.columns),
          'probability',all_probability)
pd[-2:]


# In[35]:


pd[(pd['survived']==0) &  (pd['probability']>0.9) ]


# In[36]:


pd[:5]


# In[ ]:





# In[ ]:

excel资源如下：

链接：https://pan.baidu.com/s/1PvonynplLKC6ZepSlL9DqQ
提取码：w7z3

采用多层感知器的方案的，主要是特点是针对数据的预处理过程。对excel表格的处理。

 读取文件显示前五行：

筛选出表内的指定列：

去掉名字列生成新数据：

查找未知信息null，然后补充为平均值

 

 

修改性别样式从male和female到0和1：

将DF中的数据某列拆分：

提取目标输出数据和输入数据：

将输入数据转化为0-1之间的数据方式：

随机分割数据2:8作为测试数据和训练数据的方案！：

之后建立模型，两个隐层，计算方式为上一层神经元乘下一层神经元，加偏差下一层神经元。

之后计算，绘图，预测。

插入两行数据到总数据：

 之后计算预测，找到生存概率。

筛选出实际数据为0而预测数据为存活的数据：

可以找到他们之所以没有存活的背后感人故事。。。。。。