Method forward

yolact.py:311–361

Args: - convouts (list): A list of convouts for the corresponding layers in in_channels. Returns: - A list of FPN convouts in the same order as x with extra downsample layers if requested.

(self, convouts:List[torch.Tensor])

    @script_method_wrapper
    def forward(self, convouts:List[torch.Tensor]):
        """
        Args:
            - convouts (list): A list of convouts for the corresponding layers in in_channels.
        Returns:
            - A list of FPN convouts in the same order as x with extra downsample layers if requested.
        """

        out = []
        x = torch.zeros(1, device=convouts[0].device)
        for i in range(len(convouts)):
            out.append(x)

        # For backward compatability, the conv layers are stored in reverse but the input and output is
        # given in the correct order. Thus, use j=-i-1 for the input and output and i for the conv layers.
        j = len(convouts)
        for lat_layer in self.lat_layers:
            j -= 1

            if j < len(convouts) - 1:
                _, _, h, w = convouts[j].size()
                x = F.interpolate(x, size=(h, w), mode=self.interpolation_mode, align_corners=False)
            
            x = x + lat_layer(convouts[j])
            out[j] = x
        
        # This janky second loop is here because TorchScript.
        j = len(convouts)
        for pred_layer in self.pred_layers:
            j -= 1
            out[j] = pred_layer(out[j])

            if self.relu_pred_layers:
                F.relu(out[j], inplace=True)

        cur_idx = len(out)

        # In the original paper, this takes care of P6
        if self.use_conv_downsample:
            for downsample_layer in self.downsample_layers:
                out.append(downsample_layer(out[-1]))
        else:
            for idx in range(self.num_downsample):
                # Note: this is an untested alternative to out.append(out[-1][:, :, ::2, ::2]). Thanks TorchScript.
                out.append(nn.functional.max_pool2d(out[-1], 1, stride=2))

        if self.relu_downsample_layers:
            for idx in range(len(out) - cur_idx):
                out[idx] = F.relu(out[idx + cur_idx], inplace=False)

        return out

class FastMaskIoUNet(ScriptModuleWrapper):