pytorch 模型的train模式与eval模式

原因

对于一些含有batch normalization或者是Dropout层的模型来说，训练时的froward和验证时的forward有计算上是不同的，因此在前向传递过程中需要指定模型是在训练还是在验证。

源代码

[docs]    def train(self, mode=True):
        r"""Sets the module in training mode.

        This has any effect only on certain modules. See documentations of
        particular modules for details of their behaviors in training/evaluation
        mode, if they are affected, e.g. :class:`Dropout`, :class:`BatchNorm`,
        etc.

        Returns:
            Module: self
        """
        self.training = mode
        for module in self.children():
            module.train(mode)
        return self

[docs]    def eval(self):
        r"""Sets the module in evaluation mode.

        This has any effect only on certain modules. See documentations of
        particular modules for details of their behaviors in training/evaluation
        mode, if they are affected, e.g. :class:`Dropout`, :class:`BatchNorm`,
        etc.
        """
        #该方法调用了nn.train()方法，把参数默认值改为false. 增加聚合性
        return self.train(False)

在使用含有BN层，dropout层的神经网路来说，必须要区分训练和验证