Domen Tabernik

You can use docker images that are already prebuild, but then you need to run everything inside docker which will require you to use python3.5. I am not sure how...

@akshu281, I was able to build new whl for python3.7 and python3.8. You can find them in [https://box.vicos.si/skokec/dau-convnet/](https://box.vicos.si/skokec/dau-convnet/). Binaries for python3.7 were build for TF 1.14 and TF 1.15.5, while...

@akshu281, DAU plugin was built for compatibility with TensorFlow installed from pip, therefore I used the required CUDA, cuDNN and Python version as defined in [this table on TensorFlow build...

It should be sufficent to just install cudatoolkit10xx/11xx inside the virtualenv/conda env and TensorFlow should then use this toolkit. At least it works on some my other projects in conda...

You can use the attached code ([DataAugmentator.tar.gz](https://github.com/skokec/detectron-traffic-signs/files/4545934/DataAugmentator.tar.gz)), however, this code is in matlab and is based on our raw dataset images before we exported it for the public, so you...

Hi, it is possible that the name or value in your configuration settings contains ":" which causes issue with the print_run_params function. You can just disable this function or change...

Hi, both the code and the dataset are available for research and non-commercial use only.

Yes, KolektorSDD2 is also available for research and non-commercial use only.

What are the parameters that you are passing to the `dau_conv2d` / `DAUConv2d` function ? Are you also getting the same problem if running the unit test ? You can...

That depends on the architecture where DAUs were applied. For ResNet, the DAUs replaced only 3x3 conv2d blocks inside residual blocks, while 1x1 networks remained the same. Also, for the...