v1.0.6 - Axe the color space inheritance

It can cause issues as some values were being guessed depending on the input clip's colorspace so it's best to leave it up to the user to convert instead.

setup.py

@@ -5,7 +5,7 @@
 
 setup(
     name="vsgan",
-    version="1.0.6-post1",
+    version="1.0.7",
     author="PRAGMA",
     author_email="[email protected]",
     description="VapourSynth GAN Implementation using RRDBNet, based on ESRGAN's implementation",

vsgan/__init__.py

@@ -63,12 +63,13 @@ def run(self, clip, chunk=False):
             core.std.StackVertical([results[0], results[1]]),
             core.std.StackVertical([results[2], results[3]])
         ]) if chunk else results[0]
-        # Convert back to the original color space
-        if original_format.color_family != buffer.format.color_family:
-            if original_format.color_family == vs.ColorFamily.RGB:
-                buffer = mvsfunc.ToRGB(buffer)
-            if original_format.color_family == vs.ColorFamily.YUV:
-                buffer = mvsfunc.ToYUV(buffer, css=original_format.name[3:6])
+
+        # VSGAN used to convert back to the original color space which resulted
+        # in a LOT of guessing, which was in-accurate and may not be efficient
+        # depending on what the user is doing after running VSGAN, so in all
+        # versions after 1.0.6-post1 we return it in the colorspace the GAN
+        # provides which is always RGB24.
+
         # return the new frame
         return buffer