temporal and roi_queue no use in inference phase?

[laisimiao]

temporal and roi_queue no use in inference phase?

[lizhou-cs]

I'm sorry I made a mistake while organizing the code. I have updated the new code.

JointNLT/lib/test/tracker/jointnlt.py

Lines 123 to 130 in bc8cc2f

	if self.test_method == "JOINT":
	if roi_length >= self.MAX_HIS_FRAME:
	self.roi_queue = self.roi_queue[-self.MAX_HIS_FRAME + 1:]
	roi_list = [self.template_roi_feature]
	roi_list.extend(self.roi_queue)
	temporal = torch.cat(roi_list, dim=-1)
	with torch.no_grad():
	out_dict = self.network.forward_test(self.text_dict, self.template_dict, search_patch, temporal)

[laisimiao]

okay