convert torch script(jit)

[HERIUN]

edit vgg16_bn.py. forward() *because torch jit does not supply "named tuple"

    def forward(self, X):
        h = self.slice1(X)
        h_relu2_2 = h
        h = self.slice2(h)
        h_relu3_2 = h
        h = self.slice3(h)
        h_relu4_3 = h
        h = self.slice4(h)
        h_relu5_3 = h
        h = self.slice5(h)
        h_fc7 = h
        # vgg_outputs = namedtuple(
        #     "VggOutputs", ["fc7", "relu5_3", "relu4_3", "relu3_2", "relu2_2"]
        # )
        # out = vgg_outputs(h_fc7, h_relu5_3, h_relu4_3, h_relu3_2, h_relu2_2)
        return h_fc7, h_relu5_3, h_relu4_3, h_relu3_2, h_relu2_2

convert craft to jit

craft = CRAFT()
ckpt_path = "~~"
state_dict = _get_state_dict(
            ckpt_path=ckpt_path, include="module.", delete="module.", cuda=cuda
        )
craft.load_state_dict(state_dict=state_dict, strict=True)
craft.eval()
craft = torch.jit.script(craft)

In my gpu, speed +5%(1.34 sec -> 1.27sec), memory used are same(don't know why..)

order changed(it is almost same time)