search

nicholaskajoh / ivy

Video-based object counting software.
MIT License
428 stars 170 forks source link

ENABLE_GPU_ACCELERATION is processing slow #38

Closed andikaMaulana closed 4 years ago

andikaMaulana commented 4 years ago

when i ENABLE_GPU_ACCELERATION=True, processing my video is running slow. My GPU is Nvidia RTX 2080 TI.

nicholaskajoh commented 4 years ago

Hi @andikaMaulana! Can you please quantify? How slow? You can check the logs for the processing frame rate (label: FRAME_PROCESS). Also, what detection model are you using?

By the way, did you follow the instructions in #34?

PS: Ivy doesn't run in realtime yet. But we're working on getting there.

andikaMaulana commented 4 years ago

Hi @andikaMaulana! Can you please quantify? How slow? You can check the logs for the processing frame rate (label: FRAME_PROCESS). Also, what detection model are you using?

By the way, did you follow the instructions in #34?

PS: Ivy doesn't run in realtime yet. But we're working on getting there.

I using yolo detection model.

byq-luo commented 4 years ago

Create .env from .env.example in the project root and edit as appropriate. What should I do in this step? Thank you for your guidance

nicholaskajoh commented 4 years ago

Use .env.example as an example to create a .env file. On a unix shell, you can run cp .env.example .env and then edit .env in your text editor to suit your needs.

How do you think this instruction can be better worded to make it easier to understand?

byq-luo commented 4 years ago

Use .env.example as an example to create a .env file. On a unix shell, you can run cp .env.example .env and then edit .env in your text editor to suit your needs. How do you think this instruction can be better worded to make it easier to understand?

I have solved, thank the great.

andikaMaulana commented 4 years ago

solved, I uninstall yolo34py lib, then i install yolo34py-gpu.