tensorflow实验九----------泰坦尼克号

导入库

import tensorflow as tf
import numpy as np
import matplotlib.pyplot as plt
import pandas as pd
import urllib.request
import os
%matplotlib inline
 
print("Tensorflow版本是:",tf.__version__)

data_url="http://biostat.mc.vanderbilt.edu/wili/pub/Main/DataSets/titanic3.xls"
 
data_file_path="E:\软件\微信/titanic3.xls"
 
if  not os.path.isfile(data_file_path):
    result=urllib.request.urlretrieve(data_url,data_file_path)
    print('downloaded;',result)
else:
    print(data_file_path,'data file already exists.')

data_file_path="E:\软件\微信/titanic3.xls这个是你自己要下载存放的路径

import numpy
import pandas as pd
from sklearn import preprocessing
# 读取数据文件，结果为DataFrame格式
df_data = pd.read_excel(data_file_path)

筛选提取字段

selected_cols=['survived','name','pclass','sex','age','sibsp','parch','fare','embarked']
selected_df_data=df_data[selected_cols]

def prepare_data(df_data):
    df=df_data.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)
    df['embarked']=df['embarked'].fillna('S')
    df['embarked']=df['embarked'].map({'C':0,'Q':1,'S':2}).astype(int)
    ndarray_data=df.values
    features=ndarray_data[:,1:]
    label=ndarray_data[:,0]
    minmax_scale=preprocessing.MinMaxScaler(feature_range=(0,1))
    norm_features=minmax_scale.fit_transform(features)
    return norm_features,label
```train_dataset=tf.data.Dataset.from_tensor_slices((x_train,y_train))
valid_dataset=tf.data.Dataset.from_tensor_slices((x_test,y_test))

```bash
x_data,y_data=prepare_data(selected_df_data)

train_size = int(len(x_data)*0.8)
x_train = x_data[:train_size]
y_train = y_data[:train_size]

x_test = x_data[train_size:]
y_test = y_data[train_size:]

train_dataset=tf.data.Dataset.from_tensor_slices((x_train,y_train))
valid_dataset=tf.data.Dataset.from_tensor_slices((x_test,y_test))

model=tf.keras.models.Sequential()

模型设置

model.add(tf.keras.layers.Dense(units=64,
                               input_shape=(7,),
                               use_bias=True,
                               kernel_initializer='uniform',
                               bias_initializer='zeros',
                               activation='relu'))

model.add(tf.keras.layers.Dense(units=32, 
                                activation='sigmoid'))

model.add(tf.keras.layers.Dense(units=1,
                               activation='sigmoid'))

模型参数信息

model.summary()

训练模型

model.compile(optimizer=tf.keras.optimizers.Adam(0.003),
             loss='binary_crossentropy',
             metrics=['accuracy'])

• optimizer可以是优化器的名字，如’adam’，也可以是优化器的实例
• loss是损失函数名⁻用sigmoid作为激活函数，一般损失函数选用binary_crossentropy
• binary_crossentropy⁻用softmax作为激活函数，一般损失函数选用categorical_crossentropy
• categorical_crossentropy•metrics模型要训练和评估的度量值

train_dataset=train_dataset.shuffle(buffer_size=10000)
train_dataset=train_dataset.batch(40)

train_history=model.fit(train_dataset,
                       epochs=100,
                       validation_data=train_dataset,
                       verbose=2)

准确率为80%

train_history.history

train_history.history.keys()

可视化

import matplotlib.pyplot as plt

%matplotlib inline

def visu_train_history(train_history,train_metric,validation_metric):
    plt.plot(train_history.history[train_metric])
    plt.plot(train_history.history[validation_metric])
    plt.title('Train History')
    plt.ylabel(train_metric)
    plt.xlabel('epoch')
    plt.legend(['train','validation'],loc='upper left')
    plt.show()

isu_train_history(train_history,'accuracy','val_accuracy')

visu_train_history(train_history,'loss','val_loss')

模型评估

result = model.evaluate(x=x_test,
                       y=y_test,
                       verbose=2)

model.metrics_names

selected_cols

插入数据

Jack_info=[0,'Jack',3,'male',23,1,0,5.0000,'S']
Rose_info=[1,'Rose',1,'female',20,1,0,100.0000,'S']

new_passenger_pd=pd.DataFrame([Jack_info,Rose_info],columns=selected_cols)

all_passenger_pd=selected_df_data.append(new_passenger_pd)

显示后三条数据

all_passenger_pd[-3:]

x_features,y_label=prepare_data(all_passenger_pd)

surv_probability=model.predict(x_features)

surv_probability[:5]

all_passenger_pd.insert(len(all_passenger_pd.columns),'surv_probability',surv_probability)

all_passenger_pd[-5:]

打印存活率大于90%的

all_passenger_pd[(all_passenger_pd['survived']==0)&(all_passenger_pd['surv_probability']>0.9)]

all_passenger_pd[1:5]