What does (*net[3: 5]) mean in pytorch?

See this inside the code

在这里插入图片描述

1 class ResNeXt101():
    2 def __init__(self):
        3 super(ResNeXt101, self).__init__()
        4 net = resnext101()
        # print((), net)

        5 net = list(())  # () to get the surface network of resneXt
        # for i, value in enumerate(net):
        #     print(i, value)
        6 self.layer0 = (net[:3])  # Pack the first three layers with 0, 1, 2.
        print(self.layer0)
        7 self.layer1 = (*net[3: 5])  # Packaging of layers 3 and 4
        8 self.layer2 = net[5]
        9 self.layer3 = net[6]

You can see the sixth line in the code (the serial number removes itself, I typed it in)self.layer0 = (net[:3])cap (a poem)
Seventh row.self.layer1 = (*net[3: 5])
There is a(net[:3])
respond in singing(*net[3: 5])
Not today.（）Usage, meaning, function because I don't really understand it either. Amazing then says*net[3: 5]Why do you need this?“ * ”
When code is written without the*The following problem occurs when running

在这里插入图片描述

It means that the list is not a subclass, that is, the parameters are not right

net = list(())

This line of code takes out each layer of the model to construct a list, just try printing it yourself. The approximate output is[conv()，BatchNorm2d()， ReLU，MaxPool2d]wait a minute!

在这里插入图片描述

The total is an element, not quite the same as a normal list.

When we takenet[:3]when we pass in a list of parameters, but we use the*net[:3]is passed in as a single element when the

list1 = ["conv", ("relu", "maxing"), ("relu", "maxing", 3), 3]
list2 = [list1[:1]]
list3 = [*list1[:1]]
print("list2:{}, *list1[:2]:{}".format(list1[:1], *list1[:1]))

在这里插入图片描述

The result without ✳ is a list and with ✳ is an element, so the(*net[3: 5])hit the nail on the head*net[3: 5]that is, to()Multiple layers are passed in this container.

to this article about pytorch in the (*net[3: 5]) what is the meaning of the article is introduced to this, more related pytorch (*net[3: 5]) content please search for my previous posts or continue to browse the following related articles I hope that you will support me in the future!