Bob.Liu

@experiencor We need your help...

Vote for the 2nd one. (ps: 第一个有点绿哈哈哈)

I solve my issue following @iassael advise. test GPU in GTX1080 with CUDA 8.0. review to this [site](https://github.com/baidu-research/warp-ctc/pull/48/commits/2be01008a47d092316b4cf39339aa397a0cb7e18) add these code to CMakeLists.txt ``` set(CUDA_NVCC_FLAGS "${CUDA_NVCC_FLAGS} -gencode arch=compute_53,code=sm_53") set(CUDA_NVCC_FLAGS "${CUDA_NVCC_FLAGS}...

@momenabdelkarim please update your driver. download from [here](https://www.nvidia.com/Download/index.aspx?lang=en-in)

Hi @themathgeek13 . Thanks for your detail experiments. I would like to share my experiments. Using the newest code with commit 2ca525a and using default parameter. Training in 4 TITAN...

@themathgeek13 Hi, I had using imagenet pretrained weights in this case. I will try to training from scratch and share you the result. But as you know it will take...

@dionysos4 it's not freeze backbone weights. But using different lr. You can see it in params.py file.

Hi @joaqo , The FPS is measure only include backbone. And the input size is 256x256. Please review the paper and another github.   Anyway, I had added a...

The FPS of full YOLOv3 from Paper: 

@fighting-liu Hi, which val dataset using in your testing. 5k or 40504? I would like to try it now. thanks.