Reducing RAM usage by fusing crop, resize, rescale, and quantize
In this last recipe, we will deploy the application on the Arduino Nano. However, a few extra operators are needed to recognize indoor environments with our tiny device.
In this recipe, we will learn how to fuse crop, resize, rescale, and quantize operators to reduce RAM usage. These extra operators will be needed to prepare the TFLite model's input.
The following Arduino sketch contains the code referred to in this recipe:
07_indoor_scene_recognition.ino:
Getting ready
To get ready for this recipe, we need to know what parts of the application affect RAM usage.
RAM usage is impacted by the variables allocated during the program execution, such as the input, output, and intermediate tensors of the ML model. However, the model is not solely responsible for memory utilization...
