神经网络--1.全连接未加偏置b_未加激活函数

1.代码(python)

初始化变量--->前向计算----->后向运算

此代码是初步描写，此篇博文是未加上偏置b_未加激活函数

第二篇博文是加上偏置b，_未加激活函数的代码，链接（https://blog.csdn.net/huanhuan59/article/details/104483851）

第三篇博文是加上偏置b，_加激活函数的代码，有详细介绍，链接（https://blog.csdn.net/huanhuan59/article/details/104484095）

# -*- coding: UTF-8 -*-
import numpy as np
import scipy
import pandas
import matplotlib as plt
import sklearn
import sklearn.datasets
import math
import random

"""
该数据集有3个类virginica，versicolor和setosa，每类50个样本；
每个样本是一个4维的特征向量,萼片长，萼片宽，花瓣长，花瓣宽；
['sepal length (cm)', 'sepal width (cm)', 'petal length (cm)', 'petal width (cm)']
"""
iris = sklearn.datasets.load_iris()
iris_key = iris.keys()
print(iris_key)
iris_data = iris['data']
iris_target = iris['target']
iris_target_names = iris['target_names']
iris_feature_names = iris['feature_names']
"""
print(iris_target)
print(iris_target_names)
print(iris_feature_names)
print("!finish!")
"""

W0 = np.zeros((5,4), dtype=np.float) + 1
W1 = np.zeros((4,5), dtype=np.float) + 1
W2 = np.zeros((3,4), dtype=np.float) + 1
S = np.zeros((3,1), dtype=np.float)
dS_dZ3 = np.zeros((3,1), dtype=np.float)


data_shape = np.shape(iris_data)
data_len = data_shape[0]  # 150
feature_len = data_shape[1]  # 4

# forward
for i in range(1):
    # Layer 0
    feature_vector = iris_data[i, :]
    Z_1 = np.dot(W0, feature_vector.reshape((4,1)))    # 列向量 正确
    #Z_1_1 = np.dot(W0, feature_vector.reshape((4, 1))) + b
    #Z_1_2 = np.dot(W0, feature_vector.reshape((1,4)))  # 行向量 语法错误
    #print("Z_1", Z_1)
    #print("Z_1", Z_1_1)

    # Layer 1
    Z_2 = np.dot(W1, Z_1)
    Z_2 = Z_2.reshape((4, 1))

    #Layer 2
    Z_3 = np.dot(W2, Z_2)
    Z_3 = Z_3.reshape((3, 1))

    # Layer 3
    S_sum = math.exp(Z_3[0]) + math.exp(Z_3[1]) + math.exp(Z_3[2])
    S[0] = math.exp(Z_3[0]) / S_sum
    S[1] = math.exp(Z_3[1]) / S_sum
    S[2] = math.exp(Z_3[2]) / S_sum

    if iris_target[i] == 0:
        label = np.array([1,0,0])
    if iris_target[i] == 1:
        label = np.array([0,1,0])
    if iris_target[i] == 2:
        label = np.array([0,0,1])

    Loss = -(label[0]*math.log(S[0]) + label[1]*math.log(S[1]) + label[2]*math.log(S[2]))
    #print("Loss:", Loss)



# backward
for i in range(3):
    # Layer 0
    dL_dS = 1 / S
    dL_dS = dL_dS.reshape((3, 1))
    print("dL_dS",dL_dS)

    # Layer 1
    if iris_target[i] == 0:
        dS_dZ3[0] = S[0] * (1 - S[0])
        dS_dZ3[1] = -S[0] * S[1]
        dS_dZ3[2] = -S[0] * S[2]


    if iris_target[i] == 1:
        dS_dZ3[0] = -S[0] * S[1]
        dS_dZ3[1] = S[1] * (1 - S[1])
        dS_dZ3[2] = -S[2] * S[1]

    if iris_target[i] == 2:
        dS_dZ3[0] = -S[0] * S[2]
        dS_dZ3[1] = -S[1] * S[2]
        dS_dZ3[2] = S[2] * (1 - S[2])

    dS_dZ3 = dS_dZ3.reshape((3, 1))
    print("dS_dZ3",dS_dZ3)



    # Layer 2
    dZ3_dW2 = np.tile(Z_2.reshape(1,4), (3,1))  # (3, 4)
    print("dZ3_dW2", dZ3_dW2)
    # dL_dW2 = dL_dS * dS_dZ3 * dZ3_dW2
    #        = dL_dS * dS_dZ3 * Z2_T

    dL_dW2 = dL_dS * np.dot(dS_dZ3, Z_2.reshape(1, 4))
    print("dL_dW2: ", dL_dW2)   # (3,4)


    #Layer 3
    dL_dZ2 = np.dot(W2.reshape(4,3),dL_dS*dS_dZ3)  #(4,1)
    dL_dW1 = np.dot(dL_dZ2,Z_1.reshape(1,5))

    #Layer 4
    dL_dZ1 = np.dot(W1.reshape(5, 4), dL_dZ2)  #(5,1)
    dL_dW0 = np.dot(dL_dZ1, feature_vector.reshape((1,4)))

    print("feature_vector.reshape((1,4))", feature_vector.reshape((1,4)))
    print("dL_dW1", dL_dW1)