thancaocuong

@batselem I've just take a look at your repo. But please include build system and instruction to build and run your source code. Thanks

@chenyu19880302 I've found that when use smooth L1 loss for dense displacements regression, Losses are quite small due to multiplying with dense_weight. And then, the loss values are averaged. It...

from darknet to make "shortcut" from darknet53 `'''` elif module_type == "shortcut": from_ = int(modules[i]["from"]) x = outputs[i-1] + outputs[i+from_] outputs[i] = x `'''` Now can I modify it to...

> First thanks for your excellent project, I have run it successfully and it performed best result in my task. > > Now I'm going to run it in TensorRT,...

@tijanavukovic1 you should load the model from scratch then load the pretrained state dict.

@ioir123ju could you update Readme.md to make your work more clear?

@ioir123ju currently is this correct if I say that we only can run with batchsize 1 with python binding?

@NICE-FUTURE any update on this issue? I also see the different between these 2 implementations