• Welcome to the world's largest Chinese hacker forum

    Welcome to the world's largest Chinese hacker forum, our forum registration is open! You can now register for technical communication with us, this is a free and open to the world of the BBS, we founded the purpose for the study of network security, please don't release business of black/grey, or on the BBS posts, to seek help hacker if violations, we will permanently frozen your IP and account, thank you for your cooperation. Hacker attack and defense cracking or network Security

    business please click here: Creation Security  From CNHACKTEAM

Recommended Posts

nn.Conv2d就是pytorch中定义的的一个普通的类,可以很方便的实现二维的卷积,它有五个主要的参数,

In_channels(输入通道数)out_channels(输出通道数)kernel_size(卷积核的大小)stride(卷积核移动的步长)padding(是否填充体积的张量)。

另外还有膨胀:决定是否使用空卷积,默认值为1(不使用)。这个参数的意义是卷积核上一个参数到另一个参数的距离,所以默认为1.groups:决定是否使用块卷积,默认为1。组的参数决定了输入通道被分成多少组,每组通道将被复用多少次,通过输入通道数的相应计算可知,这也要求组需要能被out_channelin_channel整除。例如,当输入和输出通道都是6,set groups=3时,那么in _ channels将被分成3组,每个out_channel将由2个in _ channels获得,每组in _ channels将被重用两次。Bias:是否加bias,默认为真。Padding_mode:即Padding的填充模式,默认为零填充。

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now