TensorFlow lite on Arduino UNO R3

Hello, I am concern about whether I can train and test a ML model of extreme less size dataset and deploy it on Arduino UNO R3. The dataset size could be around 3-4 MB.
Hope I could get answers ASAP, I am doing an project…
Thanks for your time !!!

Hi @sankalpbisan, Yes, you can able to train model. But as you mention the dataset size was extremely less which cannot guarantee that you can obtain a best performance model. For deploying the model you need to convert the trained model to tflite format and can able to deploy it. Thank You.

1 Like

Thanks for responsding !
Currently my concern is not about the accuracy, rather I just want to deploy the model on MCUs.

Hi @sankalpbisan, You can deploy it. At present these are the list of micro controllers support for tflite. Thank You

1 Like

Yeah I am aware of that.
Thanks :heart: