PNetlin.forward image normalization

[roimehrez]

    def forward(self, in0, in1):
        in0_sc = (in0 - self.shift.expand_as(in0)) / self.scale.expand_as(in0)
        in1_sc = (in1 - self.shift.expand_as(in0)) / self.scale.expand_as(in0)

        if (self.pnet_tune):
            outs0 = self.net.forward(in0)
            outs1 = self.net.forward(in1)
        else:
            outs0 = self.net[0].forward(in0)
            outs1 = self.net[0].forward(in1)

Why you don't use in0_sc to feed the net? Is it a bug or a feature?

[richzhang]

That is a bug! Thank you for the catch! I am re-running the experiments right now. The results do not change much. I'll fix the bug and release new weights.

In the meantime, you can fix the bug and retrain yourself by running bash train_test_metric.sh as well.