pytorch模型转换为coreml模型，但是报错forward() takes 1 positional argument but 2 were given

问题遇到的现象和发生背景

最近在学习ios开发，需要将一个pytorch模型转换为coreml模型，过程中需要用torch.jit.trace转换出torchscript。

问题相关代码

教程中给出的torch.jit.trace都有两个参数，如下：

    example_input = torch.rand(1, 3, 224, 224) 
    traced_model = torch.jit.trace(model, example_input)

但是modle.py中的forward()只有一个参数，如下：

   def forward(self):
        # first transfer and removal
        self.z_non_makeup_c = self.enc_content(self.non_makeup)
        self.z_non_makeup_s = self.enc_semantic(self.non_makeup_parse)
        self.z_non_makeup_a = self.enc_makeup(self.non_makeup)

        self.z_makeup_c = self.enc_content(self.makeup)
        self.z_makeup_s = self.enc_semantic(self.makeup_parse)
        self.z_makeup_a = self.enc_makeup(self.makeup)
        # warp makeup style
        self.mapX, self.mapY, self.z_non_makeup_a_warp, self.z_makeup_a_warp = self.transformer(self.z_non_makeup_c,
                                                                                                self.z_makeup_c,
                                                                                                self.z_non_makeup_s,
                                                                                                self.z_makeup_s,
                                                                                                self.z_non_makeup_a,
                                                                                                self.z_makeup_a)
        # makeup transfer and removal
        self.z_transfer = self.gen(self.z_non_makeup_c, self.z_makeup_a_warp)
        self.z_removal = self.gen(self.z_makeup_c, self.z_non_makeup_a_warp)

        # rec
        self.z_rec_non_makeup = self.gen(self.z_non_makeup_c, self.z_non_makeup_a)
        self.z_rec_makeup = self.gen(self.z_makeup_c, self.z_makeup_a)

        # second transfer and removal
        self.z_transfer_c = self.enc_content(self.z_transfer)
        # self.z_non_makeup_s = self.enc_semantic(self.non_makeup_parse)
        self.z_transfer_a = self.enc_makeup(self.z_transfer)

        self.z_removal_c = self.enc_content(self.z_removal)
        # self.z_makeup_s = self.enc_semantic(self.makeup_parse)
        self.z_removal_a = self.enc_makeup(self.z_removal)
        # warp makeup style
        self.mapX2, self.mapY2, self.z_transfer_a_warp, self.z_removal_a_warp = self.transformer(self.z_transfer_c,
                                                                                                 self.z_removal_c,
                                                                                                 self.z_non_makeup_s,
                                                                                                 self.z_makeup_s,
                                                                                                 self.z_transfer_a,
                                                                                                 self.z_removal_a)
        # makeup transfer and removal
        self.z_cycle_non_makeup = self.gen(self.z_transfer_c, self.z_removal_a_warp)
        self.z_cycle_makeup = self.gen(self.z_removal_c, self.z_transfer_a_warp)

运行结果及报错内容

运行之后报错：forward() takes 1 positional argument but 2 were given

我想要达到的结果

我现在完全不知道该怎么修改，请求帮助。

你在哪儿调用的forward啊？
example_input = torch.rand(1, 3, 224, 224)
traced_model = torch.jit.trace(model, example_input)
这两个也没调用他呀