A LA MODE的博客

前言本文比较了以下几种神经网络模型容器的区别一、容器类型1.有序型代码如下：1.1.LeNetSequentialclass LeNetSequential(nn.Module): def __init__(self, classes): super(LeNetSequential, self).__init__() self.features = nn.Sequential( nn.Conv2d(3, 6, 5),

项目场景：提示：自己搭建CNN以及ALexNet时：问题描述：有一段代码看不懂：x = x.view(x.size()[0], -1)初步分析：应该是让特征张量扁平化，一维化。class LeNetSequential(nn.Module): def __init__(self, classes): super(LeNetSequential, self).__init__() self.features = nn.Sequential(