I want to use a custom model on a raspberry pi using Google Coral accelerator. I trained the quantized mobilenet-ssd model on my dataset and I have the tflite file and the label txt file. Now I need to use edgetpu_compiler so the tflite file will be optimized for the Google Coral accelerator. The problem is that I can't install it. This is what I get (I followed these instructions: https://coral.ai/docs/edgetpu/compiler/):
pi@raspberrypi:~/Desktop/project/mobilenet_ssd_v2 $ sudo apt-get install edgetpu Reading package lists... Done Building dependency tree Reading state information... Done Some packages could not be installed. This may mean that you have requested an impossible situation or if you are using the unstable distribution that some required packages have not yet been created or been moved out of Incoming. The following information may help to resolve the situation:
The following packages have unmet dependencies: edgetpu : Depends: edgetpu-compiler but it is not installable E: Unable to correct problems, you have held broken packages.
Even when I try to specifically install edgetpu-compiler, it doesn't work:
pi@raspberrypi:~/Desktop/project/mobilenet_ssd_v2 $ sudo apt-get install edgetpu-compiler Reading package lists... Done Building dependency tree Reading state information... Done Package edgetpu-compiler is not available, but is referred to by another package. This may mean that the package is missing, has been obsoleted, or is only available from another source
E: Package 'edgetpu-compiler' has no installation candidate
Any idea how to solve this? Or is there another way to compile the tflite file?
This is because the edgetpu compiler is not supported for 32 bits OS, unfortunately. This used to be supported, but coral has moved onto only supporting 64 bits architecture. You can see the requirements here