Using tensorflow runtime on raspberry pi 4B

General Discussion
, ,
#1

Hi everyone. I am trying to build an image classification AI that classifies the different types of trash. For this I am using a Raspberry pi 4 B and a Raspberry Pi camera model 3. I have trained my AI model on Google cloud (Vertex AI) and have downloaded my file as a Tensor Flow Lite package (.tflite). I have tried to install on my raspberry pi tflite_runtime but I would always get that the wheel was not supported by the platform. I’m also having trouble with the python code I should write in order to make my AI work. Could you please help me ?
Let me know if you need additional information.

Thank you,

Constantin

#3

Hi @Constantin_Cedillo_V, If you have installed tflite_runtime using pip you will get “the wheel was not supported by the platform”. beacause those wheels installed using pip are supported for x_86 architecture. As you said you are using raspberry pi 4 model B which has an ARM architecture. so you have to build those wheels from the source. please refer to this documnet for building from source. Thank You.