Local Machine Learning Model to Edge Impulse Deployment?

#1

If I have a trained machine learning model which I trained locally on my PC and then converted to tflite is there any way I can upload that trained model (with its parameters) to edge impulse and then create an Arduino Library in the deployment section of edge impulse?

#2

Hello,

It is not supported.
Check
How would I run a pre-trained tflite model inedge impulse enviornment? and
Deploy a Pretrained Model into A Raspberry Pi

#3

@chinmaysahoo

There is a new feature, that could be of interest.
Unveiling BYOM and the Edge Impulse Python SDK
Pretrained models

#4

@chinmaysahoo Yep, just upload your model via the UI (Dashboard > Upload your model) and then deploy to Arduino library (or follow @Joeri’s links to deploy via the Python API).