tensorflow2.0 深度学习与入门实践 日月光华 学习笔记

线性回归

详情见link.

多层感知器(神经网络)

import tensorflow as tf
import pandas as pd

# 提取数据
data = pd.read_csv('/..../tfb.csv')
x = data.iloc[:,0:-1]
y = data.iloc[:,-1]

#建立顺序模型
#Dense(输出,输入,激励函数)
model = tf.keras.Sequential([tf.keras.layers.Dense(10, input_shape=(3,), activation = 'relu'),
        tf.keras.layers.Dense(1)])
#model.summary  显示模型结构
#配置优化器,损失函数
model.compile(optimizer='adam'loss='mse')
#自定义学习效率
#model.compile(optimizer=tf.keras.optimizers.Adam(learning_rate=0.01))
  
#训练模型
model.fit(x,y,epochs=100)

#预测
test = data.iloc[:10,1:-1]
model.predict(test)

逻辑回归

import tensorflow as tf
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

#数据提取
data = pd.read_csv('/..../tfb.csv')
x = data.iloc[:,:-1]
y = data.iloc[:,-1].replace(-1,0)#-1替换为0

#建立顺序模型
model = tf.keras.Sequential()
model.add(tf.keras.layers.Dense(4,input_shape=(15,),activation='relu'))
model.add(tf.keras.layers.Dense(4,activation='relu'))
model.add(tf.keras.layers.Dense(1,activation='sigmold'))

#配置优化器,损失函数
model.compile(optimizer='adam'loss='binary_crossentropy',metrics =['acc'])

# 训练模型
history = model.fit(x,y,epochs=100)

#loss以及acc图形绘制
plt.plot(history.epoch,history.history['loss'])
plt.plot(history.epoch,history.history['acc'])
plt.show()

softmax分类

import tensorflow as tf
import numpy as np
import  pandas as pd
import matplotlib.pyplot as plt

(train_image,train_lable),(test_image,test_label)=tf.keras.datasets.fashion_mnist.load_data()
#train_image.shape 显示数据形状
#plt.imshow(train_image[0])  #显示第一张图片

#数据归一化
train_image = train_image/255
test_image = test_image/255

#建立模型
model = tf.keras.Sequential()
model.add(tf.keras.layers.Flatten(input_shape=(28,28)))#降维
model.add(tf.keras.layers.Dense(128,activation='relu'))
model.add(tf.keras.layers.Dropout(0.5))#使用Dropout防止过拟合
model.add(tf.keras.layers.Dense(10,activation='softmax'))

#配置优化器,损失函数
model.compile(optimizer='adam',loss='sparse_categorical_crossentropy',metrics=['acc'])

#顺序编码转变独热编码
#train_lable_onehot = tf.keras.utils.to_catagorical(train_label)
#test_lable_onehot = tf.keras.utils.to_catagorical(test_label)
#model.compile(optimizer='adam',loss='categorical_crossentropy',metrics=['acc'])


#训练模型
model.fit(train_image,train_lable,epochs=5)
#添加训练中的acc和loss值
#model.fit(train_image,train_lable,epochs=5,validation_data=(test_image,test_lable))

#评价模型
model.evaluate(test_image,test_label)

#预测
predict = model.predict(test_image)
#具体到某一个
np.argmax(predict[0])#预测标签
test_image[0]#实际标签