outputs = (outputs + 1) / 2. in test code

[encounter1997]

Hi, I have some question about the test code:
In both test.py and test_batch.py, the decoding output was processed by the following code,
outputs = (outputs + 1) / 2.
However, this is not shown in the sample function of MUNIT_Trainer.
I wonder what is this operation designed for, and when should I use it ?

[ggt55ng6]

scale pixel value from [-1, 1] to [0, 1]