littletomatodonkey

thanks for your excellent job! it seems that the training is very slow when i use the straug(6x times slower than that without straug). What about the real speed when...

关于ACFPN中上菜样和CxAM模块的几个问题

4

你好，感谢开源ACFPN，有2个问题想请教下 1. 下面这个地方`is_upsample`应该一直设置为True?最开始P5和C4融合时，就应该对P5进行降采样（我看backbone部分只修改了dilation，应该没有对C5的分辨率进行修改？） https://github.com/Caojunxu/AC-FPN/blob/2e4e0b2d4d00f1090548f41295824d1d7997a232/detectron/modeling/FPN.py#L375 2.论文中的CxAM和CnAM模块中的1x1卷积部分，是否对通道数进行了缩减了呢？nonlocal中是对计算通道数减半，减少了计算量，不知道你是否有意愿开源这两个模块呢？  期待你的回复，多谢！

[WIP]fix trt for swin inference

1

For faster inference speed in swin-series model, you can 1. generate shape info file ```shell python3.7 python/predict_cls.py \ -c configs/inference_cls.yaml \ -o Global.batch_size=1 \ -o Global.use_tensorrt=True \ -o Global.collect_shape=True \...

fix norm on gpu

1

att

Hi, thanks for your contribtution, I am studing knowledge distillation as now, would you mind that i create a pull request of some other competitive distillation methods while using PaddlePaddle,...

kie doc

2

att.

I rebuild the code following the repo, finding the nas_fpn is too slow compared with the standard FPNm method(about 10 times slower). Is there anyone who has the similar conclusion?

about multiple object tracking

2

Hello, I was wondering whether the code can be changed to track multiple objects per frame? I tried to modify the code to let it accept multiple objects positions input...

[WIP]add reviewkd

1

att

att