Upload New File

人工智能系统实战第三期/实战代码/深度学习项目实战/神经网络实现鸢尾花分类/神经网络解决异或.py

+import numpy as np
+import numpy as np
+
+# 定义激活函数（Sigmoid）
+def sigmoid(x):
+    return 1 / (1 + np.exp(-x))
+
+# 输入数据
+X = np.array([[0, 0], [0, 1], [1, 0], [1, 1]])
+# 目标输出
+y = np.array([[0], [1], [1], [0]])
+
+# 设置随机种子，以确保结果可重复
+np.random.seed(42)
+
+# 初始化权重和偏置
+W1 = np.random.random((2, 2))
+b1 = np.zeros((1, 2))
+W2 = np.random.random((2, 1))
+b2 = np.zeros((1, 1))
+
+# 定义学习率和迭代次数
+learning_rate = 0.1
+epochs = 10000
+
+# 训练网络
+for epoch in range(epochs):
+    # 前向传播
+    hidden_layer_input = np.dot(X, W1) + b1
+    hidden_layer_output = sigmoid(hidden_layer_input)
+    # hidden_layer_output = hidden_layer_input
+    output_layer_input = np.dot(hidden_layer_output, W2) + b2
+    # output_layer_output = output_layer_input
+    output_layer_output = sigmoid(output_layer_input)
+
+    # 计算损失值
+    # loss = np.mean((output_layer_output - y) ** 2)
+
+    # 反向传播
+    output_error = (output_layer_output - y) * output_layer_output * (1 - output_layer_output)
+    hidden_layer_error = np.dot(output_error, W2.T) * hidden_layer_output * (1 - hidden_layer_output)
+
+    # 更新权重和偏置
+    W2 -= learning_rate * np.dot(hidden_layer_output.T, output_error)
+    b2 -= learning_rate * np.sum(output_error, axis=0, keepdims=True)
+    W1 -= learning_rate * np.dot(X.T, hidden_layer_error)
+    b1 -= learning_rate * np.sum(hidden_layer_error, axis=0, keepdims=True)
+
+# 用训练好的网络进行预测
+hidden_layer_input = np.dot(X, W1) + b1
+hidden_layer_output = sigmoid(hidden_layer_input)
+output_layer_input = np.dot(hidden_layer_output, W2) + b2
+output_layer_output = sigmoid(output_layer_input)
+
+# 输出每个输入对应的计算数值
+for i in range(len(X)):
+    print("输入:", X[i])
+    print("hidden_layer_input", hidden_layer_input[i])
+    print("hidden_layer_output", hidden_layer_output[i])
+    print("output_layer_input", output_layer_input[i])
+    print("output_layer_output", output_layer_output[i])
+    print("输出:", output_layer_output[i][0])
+    print()
+
+print('W1-----')
+print(W1)
+print('b1-----')
+print(b1)
+
+print('W2-----')
+print(W2)
+print('b2-----')
+print(b2)
\ No newline at end of file