Keras 使用多层感知器 预测泰坦尼克 乘客 生还概率

Posted 2020-11-10 朱小丰的丰

# coding: utf-8

# In[6]:


# -*- coding: utf-8 -*-  
import urllib.request
import os


# In[7]:


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


# In[8]:


import numpy
import pandas as pd


# In[9]:


all_df = pd.read_excel(filepath)


# In[10]:


all_df[:2]


# In[11]:


cols=[‘survived‘,‘name‘,‘pclass‘ ,‘sex‘, ‘age‘, ‘sibsp‘,
      ‘parch‘, ‘fare‘, ‘embarked‘]
all_df=all_df[cols]


# In[12]:


all_df[:2]


# In[13]:


all_df.isnull().sum()


# In[14]:


df=all_df.drop([‘name‘], axis=1)


# In[15]:


age_mean = df[‘age‘].mean()
df[‘age‘] = df[‘age‘].fillna(age_mean)


# In[16]:


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


# In[17]:


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


# In[18]:


df[:2]


# In[19]:


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


# In[20]:


x_OneHot_df[:2]


# In[21]:


ndarray = x_OneHot_df.values


# In[22]:


ndarray.shape


# In[23]:


ndarray[:2]


# In[26]:


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


# In[27]:


Features.shape


# In[28]:


Features[:2]


# In[29]:


Label.shape


# In[32]:


Label[:2]


# In[33]:


from sklearn import preprocessing


# In[34]:


minmax_scale = preprocessing.MinMaxScaler(feature_range=(0, 1))


# In[35]:


scaledFeatures=minmax_scale.fit_transform(Features)


# In[36]:


scaledFeatures[:2]


# In[37]:


Label[:5]


# In[38]:


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


# In[39]:


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


# In[40]:


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[41]:


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


# In[42]:


train_Features[:2]


# In[43]:


train_Label[:2]

# coding: utf-8

# In[1]:


import numpy
import pandas as pd
from sklearn import preprocessing
numpy.random.seed(10)


# In[2]:


all_df = pd.read_excel("data/titanic3.xls")


# In[3]:


cols=[‘survived‘,‘name‘,‘pclass‘ ,‘sex‘, ‘age‘, ‘sibsp‘,
      ‘parch‘, ‘fare‘, ‘embarked‘]
all_df=all_df[cols]


# In[4]:


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


# In[5]:


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


# In[6]:


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[7]:


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


# In[8]:


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


# In[9]:


model = Sequential()


# In[10]:


model.add(Dense(units=40, input_dim=9, 
                kernel_initializer=‘uniform‘, 
                activation=‘relu‘))


# In[11]:


model.add(Dense(units=30, 
                kernel_initializer=‘uniform‘, 
                activation=‘relu‘))


# In[12]:


model.add(Dense(units=1, 
                kernel_initializer=‘uniform‘,
                activation=‘sigmoid‘))


# In[13]:


model.compile(loss=‘binary_crossentropy‘, 
              optimizer=‘adam‘, metrics=[‘accuracy‘])


# In[14]:


train_history =model.fit(x=train_Features, 
                         y=train_Label, 
                         validation_split=0.1, 
                         epochs=30, 
                         batch_size=30,verbose=2)


# In[15]:


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()


# In[16]:


show_train_history(train_history,‘acc‘,‘val_acc‘)


# In[17]:


show_train_history(train_history,‘loss‘,‘val_loss‘)


# In[18]:


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


# In[19]:


scores[1]