SoftMax 导数计算：错误：“numpy.float64”对象不支持项目分配

Question

我正在尝试计算 SoftMax 函数的导数，但我无法调用它来消除下面提到的错误。

def softmax_grad(self,s):
    print('s.shape:',s.shape)
    jacobian_m = np.diag(s)
    print('jacobian_m:',jacobian_m.ndim)
    for i in range(len(jacobian_m)):
        print('i:',i)
        for j in range(len(jacobian_m)):
            print('j:',j)
            if i == j:
                jacobian_m[i][j] = s[i] * (1-s[i])
            else:
                jacobian_m[i][j] = -s[i]*s[j]
    return jacobian_m

def train(self, inputs, targets, eta, niterations):
    ndata = np.shape(inputs)[0] # number of data samples 
    # adding the bias
    inputs = np.concatenate((inputs,-np.ones((ndata,1))),axis=1)
    
    # numpy array to store the update weights 
    updatew1 = np.zeros((np.shape(self.weights1))) 
    updatew2 = np.zeros((np.shape(self.weights2)))
    updatew3 = np.zeros((np.shape(self.weights3)))
    
   
    for n in range(niterations):   
        # forward phase 
        self.outputs = self.forwardPass(inputs)
  
    
        # Error using the sum-of-squares error function
        error = 0.5*np.sum((self.outputs-targets)**2)
   
        if (np.mod(n,100)==0):
            print("Iteration: ",n, " Error: ",error)
    
            deltao = self.sigmoid_derivative(self.outputs)
            print('delto :',deltao)

运行以下代码后，我遇到了一个错误
TypeError: 'numpy.float64' object does not support item assignment

Answer 1

np.diag 如果参数是二维数组（在您的情况下为“s”），则返回由给定参数的对角元素组成的一维数组。这就是为什么当您尝试获取 ndim 时会收到 1。因此，您的 jacobian_m 变量是一维数组，而您正试图像使用二维数组一样使用它，这会导致错误。