Live classifications with a smartphone
When we talk of model testing, we usually refer to the evaluation of the trained model on the testing dataset. However, model testing in Edge Impulse is more than that.
In this recipe, we will learn how to test model performance on the test set and show a way to perform live classifications with a smartphone.
Getting ready
Before implementing this recipe, the only thing we need to know is how we can evaluate model performance in Edge Impulse.
In Edge Impulse, we can evaluate the trained model in two ways:
- Model testing: We assess the accuracy using the test dataset. The test dataset provides an unbiased evaluation of model effectiveness because the samples are not used directly or indirectly during training.
- Live classification: This is a unique feature of Edge Impulse whereby we can record new samples either from a smartphone or a supported device (for example, the Arduino Nano).
The live classification approach...
