一种参数、层数等可选的全连接网络构建方法（netWork.py）

"""
可选参数：
	- opt.layer_number：层数（e.g. 3）
	- opt.layer_paramete：每层参数量 （e.g. [3， 64， 64]）
	- opt.activation：none, RuLU, etc.
"""

class simpleNet(nn.Module):
    """
    定义了一个简单的三层全连接神经网络，每一层都是线性的
    """
    def __init__(self, opt):
        super(simpleNet, self).__init__()

        for i in range(0, opt.layer_number - 1):

            exec(f"self.layer{i + 1} = nn.Linear({opt.layer_parameter[i]}, {opt.layer_parameter[i + 1]})")

        exec(f"self.layer{opt.layer_number} = nn.Linear({opt.layer_parameter[opt.layer_number - 1]}, 1)")

        self.layer_number = opt.layer_number

    def forward(self, x):

        for i in range(0, self.layer_number):
            x = eval("self.layer" + str(i + 1))(x)

        return x


class Activation_Net(nn.Module):
    """
    在上面的simpleNet的基础上，在每层的输出部分添加了激活函数
    """

    def __init__(self, opt):
        super(Activation_Net, self).__init__()

        for i in range(0, opt.layer_number - 1):
            exec(f"self.layer{i + 1} = nn.Sequential(nn.Linear({opt.layer_parameter[i]}, {opt.layer_parameter[i + 1]}), nn.{opt.activation}(True))")

        exec(f"self.layer{opt.layer_number} = nn.Sequential(nn.Linear({opt.layer_parameter[opt.layer_number - 1]}, 1))")

        self.layer_number = opt.layer_number

    def forward(self, x):

        for i in range(0, self.layer_number):
            x = eval("self.layer" + str(i + 1))(x)

        return x