REFINING THE MODEL
If you find that the identification isn’t accurate, first check your model’s accuracy for that item in the Edge Impulse Model Testing tab. You can add more photos by following the Data Acquisition steps and then selecting Retrain model in Edge Impulse. You can also add more items by labeling and uploading in Data Acquisition and retraining the model. After each retraining, check for accuracy and then redeploy by running WebAssembly within Deployment.
If you want to go further with machine learning, check out Pimoroni’s new BrainCraft HAT for the Pi 4. It features a 240×240 TFT IPS display for inference output, slots for a camera connector cable, a five-way joystick, two microphones, audio outputs, and much more. Most importantly, it has a controllable fan to keep the thing cool during all the never-ending TensorFlow computations.