Giters

AlexeyAB / yolo2_light

Light version of convolutional neural network Yolo v3 & v2 for objects detection with a minimum of dependencies (INT8-inference, BIT1-XNOR-inference)

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Memory leak

AleBasso80 opened this issue · comments

commented

OS: Linux 16.04

Hello,
while running the demo I noticed that the process memory keeps on increasing.
Has anyone noticed the same issue?

Thanks
Alessandro

commented

I have that leak in case of int8 while running multiple darknets in parallel. It is fine if I disable quantization.
@AlexeyAB is there any reason for that by design?

Edit: I noticed recent commits. I will check if it is already fixed.

commented

No it is not fixed.

commented

I solved by explicitly freeing prob and mask of each element of detections.