Figure 5-5 correspondence between the variables in the code

The correspondence relationship is as follows:

Code inside dprice 1 Why is it?
Because we know the actual neural networks, back-propagation of the first item is
$\frac{\partial E}{\partial y_2}$

So here for the convenience of the reader's understanding, there is no concept of E leads.
So the meaning here is dprice $\ Frac {y_2 y_2} {}$

So the above are:
$\ Frac {} y_2 = x_2} {1.1$
$dapple\_price=\frac{\partial y_2}{\partial y_2}·\frac{\partial y_2}{\partial x_2}=1·1.1=1.1$
$dapple=\frac{\partial y_2}{\partial y_2}·\frac{\partial y_2}{\partial x_2}·\frac{\partial x_2}{\partial y_1}·\frac{\partial y_1}{\partial x_1}=1· 1.1·1·2=2.2$

Note that there is still not related to the introduction of amendments to calculate weights, so in fact and real neural networks were far worse