ML4IoTCourse

Introduction

Laboratories developed during the Machine Learning for IoT course at PoliTo.

Description

The course aims to introduce the problems related to the implementation of machine learning applications and algorithms on platforms other than high-performance servers available in the cloud. The contents and the skills acquired at the end of the course include both the hardware aspects of the problem (architectures of "edge" devices) and the software aspects (programming models, protocols and related APIs). These skills will allow a correct understanding of decentralized systems in which the flow of data is processed not only on servers, but rather locally on devices with reduced computational resources and energy.

Topics covered

1. Gather data using sensors:

   • DHT-11 (temperature and humidity),
   • USB Microfophone.
2. • Pre-Processing of Audio Signals:
     • Resampling,
     • Short Time Fourier Transform (STFT),
     • Mel-frequency cepstral coefficients (MFCCs).
   • Image pre-processing
3. • Training networks for:
     • Temperature and Humidity Forecasting
     • Keyword Spotting
   • deployment of the trained networks

4. Optimizations:

   • Post-Training Quantization (PTQ)
   • Structured Pruning via Width Scaling
   • Magnitude-Based Pruning

5. RESTfull applications

6. MQTT applications