Why does input image in XYZ while expertC image in RGB

[BeyondHeaven]

In datasets.py I can see the following:

img_input = cv2.imread(self.set1_input_files[index % len(self.set1_input_files)],-1)
img_exptC = Image.open(self.set1_expert_files[index % len(self.set1_expert_files)])

Here img_input is in XYZ while img_exptC is in RGB. But in the process of computing mse loss there is no conversion from XYZ to RGB for input image. Why does this work? In my conception the mse loss should be computed in the same color space, either XYZ or RGB.