GA优化BP网络的权值和阈值，没有优化隐含层节点的个数；还需要做进一步的改进。-Using GAs to optimize the weights and biases of BPNN, without optimazation of num.Inputs.

非线性凸优化约束问题 min 10 - x1**2 - x2**2, s.t. x1**2 - x2 <= 0 and x1 + x2 = 0(In this exercise, I use Python-Scipy to solve the following constrained nonlinear optimazation problem: min 10 - x1**2 - x2**2, s.t. x1**2 -