Error while running about deep_sort_yolov3 HOT 4 CLOSED

I think it may because of ran out of memory of your GPU while trying to allocate , there have two tf.session running one for yolov3 detect one for deep_sort track . If you use a small GPU it is easy to encounter this error . How about your GPU ?

from deep_sort_yolov3.

I am running it on TITAN X 12 GB.

from deep_sort_yolov3.

2018-05-23 11:18:50.649409: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1030] Found device 0 with properties:
name: GeForce GTX 1060 major: 6 minor: 1 memoryClockRate(GHz): 1.6705
pciBusID: 0000:01:00.0
totalMemory: 5.93GiB freeMemory: 5.57GiB
2018-05-23 11:18:50.649443: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:0) -> (device: 0, name: GeForce GTX 1060, pci bus id: 0000:01:00.0, compute capability: 6.1)
model_data/yolo.h5 model, anchors, and classes loaded.
2018-05-23 11:18:59.502250: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1120] Creating TensorFlow device (/device:GPU:0) -> (device: 0, name: GeForce GTX 1060, pci bus id: 0000:01:00.0, compute capability: 6.1)
fps= 11.191386
fps= 11.208619
fps= 11.254787

There is some message while i running it , I never encountered the error as your's , so i dont have good ideas about it . Have you monitor the GPU memory while running before the imshow window appear through “nvidia-smi” ？

from deep_sort_yolov3.

I got the error.

I have 2 GPUs. Usually tensorflow ignores the 660 GTX. This time it picked up that GPU. Works fine after masking.

from deep_sort_yolov3.