search

DoubangoTelecom / ultimateALPR-SDK

World's fastest ANPR / ALPR implementation for CPUs, GPUs, VPUs and NPUs using deep learning (Tensorflow, Tensorflow lite, TensorRT, OpenVX, OpenVINO). Multi-Charset (Latin, Korean, Chinese) & Multi-OS (Jetson, Android, Raspberry Pi, Linux, Windows) & Multi-Arch (ARM, x86).
https://www.doubango.org/webapps/alpr/
Other
593 stars 157 forks source link

AMD Ryzen 7 5700G + NVIDIA GeForce RTX 3050 FPS #265

Closed Nook2007 closed 1 year ago

Nook2007 commented 1 year ago

Hi, we have same fps with enabled and disabled gpu. This is normal? logs without gpu https://gist.github.com/Nook2007/bf19cc75c9b9dc3263b36e404857f0e9 with gpu https://gist.github.com/Nook2007/c18d29313788b45c5eec3b2fb3acc0c9

and any updates on https://github.com/DoubangoTelecom/ultimateALPR-SDK/issues/155 ? we can't start using it

DoubangoTelecom commented 1 year ago

Hi,

From your logs you're using a new GPU and a new CUDA version 11.7 while using an old Tensorflow version (1.14). This won't work (GPU will not be activated). Quote from https://github.com/DoubangoTelecom/ultimateALPR-SDK/blob/master/samples/c++/README.md#migration-to-tensorflow-2x-and-cuda-11x: "Our SDK is built and shipped with Tensorflow 1.x to make it work on oldest NVIDIA GPUs. If you want to use newest NVIDIA GPUs (e.g. RTX3060) which requires CUDA 11.x, then you'll need to upgrade the Tensorflow version. Check https://www.tensorflow.org/install/source#gpu to know which CUDA version is required for your Tensorflow version.". Follow the guide to update your Tensorflow version. If you still have issues, check the developer group's archive, there are plenty threads explaining how to fix this kind of issues. If you still don't have GPU acceleration, then open a new thread on the dev-group instead of using the issue tracker.

DoubangoTelecom commented 1 year ago

Please notice at https://www.tensorflow.org/install/source#gpu that even the latest Tensorflow (2.9) only requires CUDA 11.2 while you're using CUDA 11.7. Our guide fir migration recommend Tensorflow 2.6 which also needs CUDA 11.2. Tensorflow 2.6 may work with your CUDA version but if it doesn't you'll have to downgrade it.