About training details

[fuenwang]

Hi, thanks for contributing this great work and dataset!
I would like to implement a pytorch version and some training details seems missing. Could you provide me the details like

1. batch size
2. the parameter of RMSProp (momentum, weight_decay, .....)
3. Data augmentation
4. Model initialization (xavier or uniform....)
5. Is your conv/deconv layer including batchnorm layer?
6. In your python code, you define " decay_heatmap(heatmap, sigma2=4):", in which sigma is 4, but the sigma in the paper is 2 instead. Which one is the final value you use for training?

Thanks very much!
Bests.