I believe the original paper means that we want the source image to mimic the color performance of the target image, and output a transferred source image. However, in your implementation, the target image is instead transferred. This is what makes the following formulas inconsistent with the paper.
I believe the original paper means that we want the source image to mimic the color performance of the target image, and output a transferred source image. However, in your implementation, the target image is instead transferred. This is what makes the following formulas inconsistent with the paper.
https://github.com/jrosebr1/color_transfer/blob/6724ccf927f2e3838ccd3f3da8b2542b79c9bf2f/color_transfer/__init__.py#L50-L68