sergiomsilva / alpr-unconstrained

License Plate Detection and Recognition in Unconstrained Scenarios
Other
1.72k stars 607 forks source link

gpu is slow #122

Open roysG opened 4 years ago

roysG commented 4 years ago

the detect take 7sec for 1 image on gpu this is a lot of time. can we make it lower? Maybe there is also option to converted to android app?

Also my card graphic info

nvidia-smiFri Mar 20 13:35:33 2020
+-----------------------------------------------------------------------------+ | NVIDIA-SMI 440.64.00 Driver Version: 440.64.00 CUDA Version: 10.2 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | |===============================+======================+======================| | 0 GeForce GTX 1070 Off | 00000000:01:00.0 On | N/A | | 32% 28C P8 7W / 151W | 360MiB / 8116MiB | 0% Default | +-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+ | Processes: GPU Memory | | GPU PID Type Process name Usage | |=============================================================================| | 0 1031 G /usr/lib/xorg/Xorg 203MiB | | 0 1547 G compiz 153MiB | +-----------------------------------------------------------------------------+

MingRongXi commented 3 years ago

Hi, roysG! I encountered the same problem, did you solve it?

AnhHC commented 3 years ago

You can disable eager execution on Keras following this https://stackoverflow.com/questions/53429896/how-do-i-disable-tensorflows-eager-execution

This will speed up the inference time.

Thanks