Why some conv layers is_training was False.

[logic1988]

Why the is_training param of two conv layers was set to False:
https://github.com/mgharbi/hdrnet_legacy/blob/master/hdrnet/models.py#L116
https://github.com/mgharbi/hdrnet_legacy/blob/master/hdrnet/models.py#L131
Any influence on result?

[mgharbi]

That should not matter, those layers have no batchnorm. is_training switches between using the running mean/std (at train time) vs. using the memorized values (at test time).