PyTorch模型定义
1. 前言
好久没更新了,2022年也过去快一半了,更文量还是不如前几年。近期会尝试加快更新的进度。
本篇文章的更新内容是PyTorch模型的定义。
2. PyTorch模型定义的方式
2.1. Sequential
Sequential 类可以通过更加简单的方式定义模型。它可以接收一个子模块的有序字典(OrderedDict) 或者一系列子模块作为参数来逐一添加 Module 的实例,⽽模型的前向计算就是将这些实例按添加的顺序逐⼀计算
Sequential定义源码:
torch.nn.modules.container—PyTorch1.11.0documentationclassSequential(Module): r"""A sequential container. Modules will be added to it in the order they are passed in the constructor. Alternatively, an ``OrderedDict`` of modules can be passed in. The ``forward()`` method of ``Sequential`` accepts any input and forwards it to the first module it contains. It then "chains" outputs to inputs sequentially for each subsequent module, finally returning the output of the last module. The value a ``Sequential`` provides over manually calling a sequence of modules is that it allows treating the whole container as a single module, such that performing a transformation on the ``Sequential`` applies to each of the modules it stores (which are each a registered submodule of the ``Sequential``). What's the difference between a ``Sequential`` and a :class:`torch.nn.ModuleList`? A ``ModuleList`` is exactly what it sounds like--a list for storing ``Module`` s! On the other hand, the layers in a ``Sequential`` are connected in a cascading way. Example:: # Using Sequential to create a small model. When `model` is run, # input will first be passed to `Conv2d(1,20,5)`. The output of # `Conv2d(1,20,5)` will be used as the input to the first # `ReLU`; the output of the first `ReLU` will become the input # for `Conv2d(20,64,5)`. Finally, the output of # `Conv2d(20,64,5)` will be used as input to the second `ReLU` model = nn.Sequential( nn.Conv2d(1,20,5), nn.ReLU(), nn.Conv2d(20,64,5), nn.ReLU() ) # Using Sequential with OrderedDict. This is functionally the # same as the above code model = nn.Sequential(OrderedDict([ ('conv1', nn.Conv2d(1,20,5)), ('relu1', nn.ReLU()), ('conv2', nn.Conv2d(20,64,5)), ('relu2', nn.ReLU()) ])) """_modules: Dict[str, Module] # type: ignore[assignment]def__init__(self, *args: Module) ->None: ... def__init__(self, arg: 'OrderedDict[str, Module]') ->None: ... def__init__(self, *args): super(Sequential, self).__init__() iflen(args) ==1andisinstance(args[0], OrderedDict): forkey, moduleinargs[0].items(): self.add_module(key, module) else: foridx, moduleinenumerate(args): self.add_module(str(idx), module)
以上是节选的源码。
重点需要看的代码区域是:
def__init__(self, *args): super(Sequential, self).__init__() iflen(args) ==1andisinstance(args[0], OrderedDict): forkey, moduleinargs[0].items(): self.add_module(key, module) else: foridx, moduleinenumerate(args): self.add_module(str(idx), module)
由python基础知识可以知道,*args
代表输入的参数可以是列表
所以,这个构造函数中的if len(args) == 1 and isinstance(args[0], OrderedDict):
用来判断输入的参数是不是一个列表:
第一个判断条件是args
的长度是否为1
第二个判断条件是isinstance(args[0], OrderedDict)
,判断传入的是不是一个OrderedDict
再次:
如果不是以上的情况,那么传入的就是一些Module,接着继续处理。
使用Sequential来定义模型。只需要将模型的层按序排列起来即可,根据层名的不同,排列的时候有两种方式:
2.1.1 使用OrderedDict
对应源码if len(args) == 1 and isinstance(args[0], OrderedDict):
判断语句为真。
importcollectionimporttorch.nnasnnnet2=nn.Sequential(collections.OrderedDict([ ('fc1', nn.Linear(784, 256)), ('relu1', nn.ReLU()), ('fc2', nn.Linear(256, 10)) ])) print(net2)
使用Sequential
定义模型的好处在于简单、易读,同时使用Sequential
定义的模型不需要再写forward
,因为顺序已经定义好了。但使用Sequential
也会使得模型定义丧失灵活性,比如需要在模型中间加入一个外部输入时就不适合用Sequential
的方式实现。使用时需根据实际需求加以选择。
2.1.2 直接排列
对应源码if len(args) == 1 and isinstance(args[0], OrderedDict):
判断语句为假,进入else
语句部分
importtorch.nnasnnnet=nn.Sequential( nn.Linear(784, 256), nn.ReLU(), nn.Linear(256, 10), ) print(net)
2.2. ModuleList
对应模块为nn.ModuleList()
部分源码:
classModuleList(Module): r"""Holds submodules in a list. :class:`~torch.nn.ModuleList` can be indexed like a regular Python list, but modules it contains are properly registered, and will be visible by all :class:`~torch.nn.Module` methods. Args: modules (iterable, optional): an iterable of modules to add Example:: class MyModule(nn.Module): def __init__(self): super(MyModule, self).__init__() self.linears = nn.ModuleList([nn.Linear(10, 10) for i in range(10)]) def forward(self, x): # ModuleList can act as an iterable, or be indexed using ints for i, l in enumerate(self.linears): x = self.linears[i // 2](x) + l(x) return x """_modules: Dict[str, Module] # type: ignore[assignment]def__init__(self, modules: Optional[Iterable[Module]] =None) ->None: super(ModuleList, self).__init__() ifmodulesisnotNone: self+=modules
可以看到,ModuleList
这个类实际上功能实现的比较简单。
ModuleList
接收一个子模块(或层,需属于nn.Module
类)的列表作为输入,然后也可以类似List
那样进行append
和extend
操作。同时,子模块或层的权重也会自动添加到网络中来。
net=nn.ModuleList([nn.Linear(784, 256), nn.ReLU()]) net.append(nn.Linear(256, 10)) # # 类似List的append操作print(net[-1]) # 类似List的索引访问print(net)
nn.ModuleList 并没有定义一个网络,它只是将不同的模块储存在一起。ModuleList中元素的先后顺序并不代表其在网络中的真实位置顺序,需要经过forward函数指定各个层的先后顺序后才算完成了模型的定义。具体实现时用for循环即可完成.
classmodel(nn.Module): def__init__(self, ...): super().__init__() self.modulelist= ... ... defforward(self, x): forlayerinself.modulelist: x=layer(x) returnx
2.3. ModuleDict
对应模块为nn.ModuleDict()
ModuleDict
和ModuleList
的作用类似,只是ModuleDict
能够更方便地为神经网络的层添加名称。
部分源码:
classModuleDict(Module): r"""Holds submodules in a dictionary. :class:`~torch.nn.ModuleDict` can be indexed like a regular Python dictionary, but modules it contains are properly registered, and will be visible by all :class:`~torch.nn.Module` methods. :class:`~torch.nn.ModuleDict` is an **ordered** dictionary that respects * the order of insertion, and * in :meth:`~torch.nn.ModuleDict.update`, the order of the merged ``OrderedDict``, ``dict`` (started from Python 3.6) or another :class:`~torch.nn.ModuleDict` (the argument to :meth:`~torch.nn.ModuleDict.update`). Note that :meth:`~torch.nn.ModuleDict.update` with other unordered mapping types (e.g., Python's plain ``dict`` before Python version 3.6) does not preserve the order of the merged mapping. Args: modules (iterable, optional): a mapping (dictionary) of (string: module) or an iterable of key-value pairs of type (string, module) Example:: class MyModule(nn.Module): def __init__(self): super(MyModule, self).__init__() self.choices = nn.ModuleDict({ 'conv': nn.Conv2d(10, 10, 3), 'pool': nn.MaxPool2d(3) }) self.activations = nn.ModuleDict([ ['lrelu', nn.LeakyReLU()], ['prelu', nn.PReLU()] ]) def forward(self, x, choice, act): x = self.choices[choice](x) x = self.activations[act](x) return x """_modules: Dict[str, Module] # type: ignore[assignment]def__init__(self, modules: Optional[Mapping[str, Module]] =None) ->None: super(ModuleDict, self).__init__() ifmodulesisnotNone: self.update(modules)
实例:
net=nn.ModuleDict({ 'linear': nn.Linear(784, 256), 'act': nn.ReLU(), }) net['output'] =nn.Linear(256, 10) # 添加print(net['linear']) # 访问print(net.output) print(net)
3. 方式的区别
Sequential
适用于快速验证结果,因为已经明确了要用哪些层,直接写一下就好了,不需要同时写__init__
和forward
;
ModuleList
和ModuleDict
在某个完全相同的层需要重复出现多次时,非常方便实现,可以”一行顶多行“;
当我们需要之前层的信息的时候,比如 ResNets
中的残差计算,当前层的结果需要和之前层中的结果进行融合,一般使用 ModuleList/ModuleDict
比较方便。
参考资料
5.1 PyTorch模型定义的方式 — 深入浅出PyTorch (datawhalechina.github.io)
Sequential — PyTorch 1.11.0 documentation
torch.nn.modules.container — PyTorch 1.11.0 documentation
ModuleList — PyTorch 1.11.0 documentation
torch.nn.modules.container — PyTorch 1.11.0 documentation