神经网络中的溢出|蟒蛇

Question

我已经建立了神经元网络，但出现了溢出错误，我该如何解决？ 这是一个神经网络，我们为其提供固定值，并学会在两个输出中返回0和1。 在这种情况下，它具有4层，一个输入层具有2个神经元，两个隐藏层具有5个神经元，而输出层则具有两个神经元。拓扑列表给出的形状

import numpy as np

class Layer:
    activation_func = None
    bias = None
    weights = None
    
    def __init__(self, inputs, neurons, activation_func):
        self.activation_func = activation_func
        self.weights = np.random.rand(neurons, inputs) * 2 - 1
        self.bias = np.random.rand(1, neurons) * 2 - 1

class Net:
    layers = []
    activation_func = None
    activation_func_derivate = None
    output = []
    adds = []
    inputs = []
    
    def __init__(self, inputs, topology, activation_func, activation_func_derivate):
        #intputs: inputs number by neuron in the input layer
        #topology: each element it's the neurons number of each layer. Topology lenght it's the layers number 
        self.activation_func = activation_func
        self.activation_func_derivate = activation_func_derivate
        layers = []
        for i in range(len(topology)):
            if i == 0:
                layers.append(Layer(inputs, topology[i], self.activation_func))
            else:
                layers.append(Layer(topology[i - 1], topology[i], self.activation_func))
        self.layers = layers
    
    def forward(self, inputs):
        output = []
        adds = []
        self.inputs = inputs
        for i, l in enumerate(self.layers):
            if i == 0:
                output_aux = [[]]
                adds_aux = [[]]
                for j, x in enumerate(self.inputs):
                    z = l.weights[j]@x.T + l.bias[0][j]
                    adds_aux[0].append(z[0])
                    act = self.activation_func(z[0])
                    output_aux[0].append(act)
                output.append(np.array(output_aux))
                adds.append(np.array(adds_aux))
            else:
                z = output[i - 1]@l.weights.T + l.bias
                adds.append(z)
                act = self.activation_func(z)
                output.append(act)
        self.output = output
        self.adds = adds
        return output[len(output) - 1]
    
    def backpropagation(self, output_expected, cost_func_derivate, ratio = 0.8):
        deltas = []
        w_derivates = []
        b_derivates = []
        for i in range(len(self.output) - 1, - 1, -1):
            if i == len(self.output) - 1:
                delta = cost_func_derivate(self.output[i], output_expected)*self.activation_func_derivate(self.output[i])
                w_der = self.output[i - 1].T@delta
            else:
                act_func_der = self.activation_func_derivate(self.adds[i])
                delta = (deltas[0]@self.layers[i + 1].weights)*act_func_der
                w_der = self.output[i - 1].T@delta
            deltas.insert(0, delta)
            w_derivates.insert(0, w_der)
            b_derivates.insert(0, delta)
        
        for i, l in enumerate(self.layers):
            l.weights = l.weights - ratio*w_derivates[i].T
            l.bias = l.bias - ratio*b_derivates[i]
         
sigm = lambda x: 1/(1 + np.e**(-x))
sigm_derivate = lambda x: x*(1 - x) # x: activation
cost_func_derivate = lambda y, ye: (y - ye) #y: output layer

topology = [2, 5, 5, 2] 
inputs_net = 2
inputs_test = [np.array([[0.56, 0.75]]), np.array([[0.23, 0.41]])]
output_expected = np.array([[1, 0]])
net = Net(inputs_net, topology, sigm, sigm_derivate)

for i in range(0, 40):
    result = net.forward(inputs_test)
    print(result)
    net.backpropagation(output_expected, cost_func_derivate)

[[0.49781526 0.43428713]]
[[0.54692239 0.38037559]]
[[0.59013316 0.33654922]]
[[0.62697339 0.30091109]]
[[0.6578928  0.27148578]]
[[0.68372401 0.24666191]]
...
[[0.81010194 0.15138157]]
[[0.81775565 0.14460523]]
[[0.82502874 0.13799208]]
[[0.8327054  0.13172268]]
[[0.84106426 0.12684122]]
[[nan nan]]
[[nan nan]]
...
[[nan nan]]
[[nan nan]]
[[nan nan]]
C:/***/Test2.py:78: RuntimeWarning: overflow encountered in double_scalars
  
C:/***/Test2.py:78: RuntimeWarning: overflow encountered in power
  
C:/***/Test2.py:79: RuntimeWarning: overflow encountered in multiply
  
C:/***/Test2.py:69: RuntimeWarning: invalid value encountered in matmul
  delta = (deltas[0]@self.layers[i + 1].weights)*act_func_der