Can rawpy do postprocessing raw data in type of ndarray?

[leonmakise]

I read those issuses and know that it is impossiable to save a .arw file.

But I train a network to input a 1 channel raw data and output a 1 channel raw data. The input data in processed by .raw_image_visible.astype(np.float32) to get ndarray type. However, I cannot do postprocess in RawPy due to it is not 'rawpy.RawPy' type.

Do you convert 'rawpy.RawPy' type to any numerical type in method '.postprocess'? Can I realize those postprocessing by using parts of this repo?

Thanks for your attention and valuable time!

[kmilos]

You should be able to write back to RawPy.raw_image_visible before calling RawPy.postproceess(), please see the documentation: "Modifying the returned array directly influences the result of calling postprocess()."

But you have to make sure it is uint16 and the same scaling as the original data.