ValueError：形狀（56,1）和（56,2）未對齊：1（dim 1）！= 56（dim 0）

Question

嘗試按如下方式運行'train'函數時出現值錯誤。如何解決點積問題？

def train(self, inputs_list, targets_list): 
     # Convert inputs list to 2d array 
     inputs = np.array(inputs_list, ndmin=2).T 
     targets = np.array(targets_list, ndmin=2).T 

     #### Implement the forward pass here #### 
     ### Forward pass ### 
     # TODO: Hidden layer 
     hidden_inputs = np.dot(inputs,self.weights_input_to_hidden.T) 
     # signals into hidden layer 
     hidden_outputs = self.sigmoid(hidden_inputs) 
     # signals from hidden layer 
     print(hidden_outputs) 

     # TODO: Output layer 
     final_inputs = np.dot(hidden_outputs,self.weights_hidden_to_output) 
     # signals into final output layer 
     final_outputs = final_inputs 
     # signals from final output layer 

     #### Implement the backward pass here #### 
     ### Backward pass ### 

     # TODO: Output error 
     output_errors = final_outputs - targets_list 
     # Output layer error is the difference between desired target and actual output. 

     # TODO: Backpropagated error 
     hidden_errors = np.dot(output_errors,self.weights_hidden_to_output) 
     # errors propagated to the hidden layer 
     hidden_grad = hidden_outputs * (1 - hidden_outputs) 
     # hidden layer gradients 

     # TODO: Update the weights 
     self.weights_hidden_to_output += self.lr * output_errors * hidden_outputs 
     # update hidden-to-output weights with gradient descent step 
     self.weights_input_to_hidden += self.lr * hidden_errors * hidden_grad * inputs 
     # update input-to-hidden weights with gradient descent step 


<ipython-input-21-c3bea2c48af8> in train(self, inputs_list, targets_list) 
    31   ### Forward pass ### 
    32   # TODO: Hidden layer 
---> 33   hidden_inputs = np.dot(inputs,self.weights_input_to_hidden.T) 
    34   # signals into hidden layer 
    35   hidden_outputs = self.sigmoid(hidden_inputs) 

ValueError: shapes (56,1) and (56,2) not aligned: 1 (dim 1) != 56 (dim 0) 

Answer 1

檢查輸入和weights_input_to_hidden的形狀。我相信你可能不需要轉置。

Answer 2

，所以你只需要改變 hidden_​​inputs = np.dot（輸入，self.weights_input_to_hidden.T） 到 hidden_​​inputs = np.dot（self.weights_input_to_hidden，輸入） ？