have anyone implemented the inference on the CPU without GPU? when i run the code on the CPU, it occurs follow the errors "AssertionError: Torch not compiled with CUDA enabled"

Zhongdao / Towards-Realtime-MOT

Joint Detection and Embedding for fast multi-object tracking

MIT License

2.38k stars 539 forks source link

have anyone implemented the inference on the CPU without GPU? when i run the code on the CPU, it occurs follow the errors "AssertionError: Torch not compiled with CUDA enabled" #131

Open cverzq opened 4 years ago

cverzq commented 4 years ago

@Zhongdao hello, according to your requirement，, I have deleted the "Cuda()" function in related files. However, it also occurs the problem. could you explain to me and how to solve the problem, thank you, hope to receive your answer. moterror

cverzq commented 4 years ago

I use command "python track.py --cfg cfg\yolov3_1088x608.cfg --weights weights\latest.pt --test-mot16"

Veronica1997 commented 4 years ago

I meet the same problem with you , have you solved this problem yet?

ks322181 commented 4 years ago

I have met the same problem