Some questions about the supervise loss

[JiangHai0929]

Hi, I have some questions about the supervise loss as follows; _ss_loss_o, ss_stats_o = self.objective(estimated_flow_target_prime_to_target_directly, mini_batch['flow_map'], mask=minibatch['mask'], the mini_batch['flow_map'] is the ground-truth flow you generate in the online_triplet_creation.py (_flow_gt = self.synthetic_flow_generator(mini_batch=minibatch, training=training, net=net)) and the target image prime is warped by using the flow_gt, so the mini_batch['flow_map'] should be _flow_target_to_target_primedirectly, why you caclulate the L1 distance between _estimated_flow_target_prime_to_target_directly and mini_batch['flowmap'] rather than the distance between _estimated_flow_target_to_target_prime_directly and mini_batch['flowmap']. Thanks, Looking forward to your reply!

[PruneTruong]

Hey, so I assume you answered your question? :)

[JiangHai0929]

yes, i made a mistake, thank you!