MagicSource

@leon-liangwu Thanks for replying.... I have one more question still. You set class to 1, but, in our data, we should have 80 classes stored? How can do specific the...

If change create_dataxxx.py to select multi classes such as car and person, and also edit classes num in prototxt, would that able to work?

@leon-liangwu But if using classes more than person, then it should not have keypoints, only box and mask, does it compatible to using kps_data_layer to load data?

@leon-liangwu I have tried mask training, I can not reproduce your effect. 40000 iterations for step1 and 40000 iterations for step2: 

After changes `prop_num` to 128, larged batchsize, I still can not get replicated result of mask. the result shows only detection boxes but I can not see any mask.... 

@marekjaszuk Have u got any mask result? What did u changed? Original prototxt definitely can not produces good result except you change somewhere like prop_num

@marekjaszuk That's wired, I using prop_num=128 and batch_size=2 Why batch_size so effect result? It shouldn't be... Did u trained step1 40000 and step2 40000?

Batchsize 32 also out of memory on GTX1080ti..,..

@marekjaszuk How do u able to using multi-gpu training? Does it support nccl?

OK, I am able build caffe with nccl and training with multi-gpu support. I am setting batchsize maxium 20 and prop_num 128