YZ

the implementation of the yolov3

3

i read the code, but i found that the loss function is ssd's? the multibox loss function is't the function in yolo's paper, right?

can't find the preprocess of the data

2

like the title, i read the code but i can't find the part of the preprocess of the data, any suggestions would be appreciate!

the batch didn't slip when use muti-gpu

1

when i use two gpu and set the batch_size=16, i found that the batch is 16 on per gpu, not 8, why?

why set the alpha to 0.1？

1

https://github.com/tiancity-NJU/da-faster-rcnn-PyTorch/blob/47cd8a80f4811a504d4cb57d2b21401ccd2b1151/lib/model/da_faster_rcnn/DA.py#L29 with the alpha=0.1 and arg.lamda=0.1, the total scale seem to be 0.01?

is there a efficient way to shuffle the data?

4

dataset = dataset.repeat().shuffle(70000).batch(batch_size).prefetch(batch_size) i test the shuffle function and i believe the buffer_size decide the max index of the original data can be sampled, and my data is huge, so...

should it set to be 0.5?or 0.3 can boost the performance?

thanks for your work, like the title, is there any improvements?

can't run the released binary version under linux

17

when i run the CityEnviron.sh in ubuntu18.04 i got this error: Increasing per-process limit of core file size to infinity. Signal 11 caught. Malloc Size=131076 LargeMemoryPoolOffset=131092 CommonLinuxCrashHandler: Signal=11 Malloc Size=65535...