search

mosheliv / tensortrt-yolo-python-api

Python api for tensorrt implementation of yolov2
33 stars 5 forks source link

Wrapper crashes when two different yolo model used in the same memory space #7

Open DogukanAltay opened 4 years ago

DogukanAltay commented 4 years ago

Basically I am trying to load 2 different tiny-yolov3 models to memory and do inference depending on the situation. However, when I try to load 2 different yolo models in the same python script, library messes up the memory locations and fails.

Any help would be appreciated. I will post if I come up with a solution

mosheliv commented 4 years ago

I have never tried it like with two models. One simple solution would be to create a subprocess with the requested model (or even two) and use queues to communicate with these. It will also be much more efficient as both models csn run together and in parallel. I am not actively maintaining this repo, i am actually using the mobilnet ssd implementation aastal wrote.

On Fri, Jan 10, 2020, 00:02 DogukanAltay notifications@github.com wrote:

Basically I am trying to load 2 different tiny-yolov3 models to memory and do inference depending on the situation. However, when I try to load 2 different yolo models in the same python script, library messes up the memory locations and fails.

Any help would be appreciated. I will post if I come up with a solution

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub https://github.com/mosheliv/tensortrt-yolo-python-api/issues/7?email_source=notifications&email_token=AC7IWCY7ZEO2JUODM6D6QLDQ4375HA5CNFSM4KEWOGCKYY3PNVWWK3TUL52HS4DFUVEXG43VMWVGG33NNVSW45C7NFSM4IFAYYIQ, or unsubscribe https://github.com/notifications/unsubscribe-auth/AC7IWC4CZWTYJTK3DYJSFALQ4375HANCNFSM4KEWOGCA .

DogukanAltay commented 4 years ago

Suggested approach is the logical one however for my requirements, sugessted approach is not applicable. Thanks for the work and help.