Data Science with the Raspberry Pi: Real-Time Applications Using a Localized Cloud

Implement real-time data processing applications on the Raspberry Pi. This book uniquely helps you work with data science concepts as part of real-time applications using the Raspberry Pi as a localized cloud. You’ll start with a brief introduction to data science followed by a dedicated look at the fundamental concepts of Python programming. Here you’ll install the software needed for Python programming on the Pi, and then review the various data types and modules available. The next steps are to set up your Pis for gathering real-time data and incorporate the basic operations of data science related to real-time applications. You’ll then combine all these new skills to work with machine learning concepts that will enable your Raspberry Pi to learn from the data it gathers. Case studies round out the book to give you an idea of the range of domains where these concepts can be applied. By the end of Data Science with the Raspberry Pi, you’ll understand that many applications are now dependent upon cloud computing. As Raspberry Pis are cheap, it is easy to use a number of them closer to the sensors gathering the data and restrict the analytics closer to the edge. You’ll find that not only is the Pi an easy entry point to data science, it also provides an elegant solution to cloud computing limitations through localized deployment. What You Will Learn Interface the Raspberry Pi with sensors Set up the Raspberry Pi as a localized cloud Tackle data science concepts with Python on the Pi Who This Book Is For Data scientists who are looking to implement real-time applications using the Raspberry Pi as an edge device and localized cloud. Readers should have a basic knowledge in mathematics, computers, and statistics. A working knowledge of Python and the Raspberry Pi is an added advantage. Table of Contents About the Authors About the Technical Reviewer Acknowledgments Introduction Chapter 1: Introduction to Data Science Importance of Data Types in Data Science Data Science: An Overview Data Requirements Data Acquisition Data Preparation Data Processing Data Cleaning Duplicates Human or Machine Errors Missing Values Outliers Transforming the Data Data Visualization Data Analysis Modeling and Algorithms Report Generation/Decision-Making Recent Trends in Data Science Automation in Data Science Artificial Intelligence–Based Data Analyst Cloud Computing Edge Computing Natural Language Processing Why Data Science on the Raspberry Pi? Chapter 2: Basics of Python Programming Why Python? Python Installation Python IDEs PyCharm Spyder Jupyter Notebook Python Programming with IDLE Python Comments Python Data Types Numeric Data Types int float complex bool Numeric Operators Sequence Data Types list tuple str set dict Type Conversion Control Flow Statements if Statement if-else Statement if...elif...else statement while loop for loop Exception Handling Functions Python Libraries for Data Science NumPy and SciPy for Scientific Computation Scikit-Learn for Machine Learning Pandas for Data Analysis TensorFlow for Machine Learning Chapter 3: Introduction to the Raspberry Pi What Can You Do with the Raspberry Pi? Physical Computing with the Raspberry Pi How to Program the Raspberry Pi? Raspberry Pi Hardware System on a Chip Raspberry Pi RAM Connectivity Setting Up the Raspberry Pi microSD Memory Card Installing the OS Inserting the microSD Memory Card Connecting a Keyboard and Mouse Connecting a Monitor Powering the Raspberry Pi Raspberry Pi Enclosure Raspberry Pi Versions Raspberry Pi 1 Raspberry Pi 2 Raspberry Pi 3 Raspberry Pi Zero (W/WH) Raspberry Pi 4 Recommended Raspberry Pi Version Interfacing the Raspberry Pi with Sensors GPIO Pins GPIO Pinout GPIO Outputs Controlling GPIO Output with Python GPIO Input Signals Reading GPIO Inputs with Python Digital Signals from Sensors Analog Signals from Sensors Interfacing a Ultrasonic Sensor with the Raspberry Pi Interfacing the Temperature and Humidity Sensor with the Raspberry Pi Interfacing the Soil Moisture Sensor with the Raspberry Pi Interfacing Cameras with the Raspberry Pi Raspberry Pi as an Edge Device Edge Computing in Self-Driving Cars What Is an Edge Device? Edge Computing with the Raspberry Pi Raspberry Pi as a Localized Cloud Cloud Computing Raspberry Pi as Localized Cloud Connecting an External Hard Drive Connecting USB Accelerator Chapter 4: Sensors and Signals Signals Analog and Digital Signals Continuous-Time and Discrete-Time Signals Deterministic and Nondeterministic Signals One-Dimensional, Two-Dimensional, and Multidimensional Signals Gathering Real-Time Data Data Acquisition Sensors Analog Sensors Digital Sensors What Is Real-Time Data? Real-Time Data Analytics Getting Real-Time Distance Data from an Ultrasonic Sensor Interfacing an Ultrasonic Sensor with the Raspberry Pi Getting Real-Time Image Data from a Camera Getting Real-Time Video from a Webcam Getting Real-Time Video from Pi-cam Data Transfer Serial and Parallel Communication Interfacing an Arduino with the Raspberry Pi Serial via USB Serial via GPIOs Data Transmission Between an Arduino and the Raspberry Pi Arduino Code Raspberry Pi Python Code Time-Series Data Time-Series Analysis and Forecasting Memory Requirements More Storage More RAM Case Study: Gathering the Real-Time Industry Data Storing Collected Data Using Pandas Dataframes Saving Data as a CSV File Saving as an Excel File Reading Saved Data Files Adding the Date and Time to the Real-Time Data Industry Data from the Temperature and Humidity Sensor Chapter 5: Preparing the Data Pandas and Data Structures Installing and Using Pandas Pandas Data Structures Series DataFrame Reading Data Reading CSV Data Reading Excel Data Reading URL Data Cleaning the Data Handling Missing Values Handling Outliers Z-Score Filtering Out Inappropriate Values Removing Duplicates Chapter 6: Visualizing the Data Matplotlib Library Scatter Plot Line Plot Histogram Bar Chart Pie Chart Other Plots and Packages Chapter 7: Analyzing the Data Exploratory Data Analysis Choosing a Dataset Modifying the Columns in the Dataset Statistical Analysis Uniform Distribution Binomial Distribution Normal Distribution Statistical Analysis of Boston Housing Price Dataset Chapter 8: Learning from Data Forecasting from Data Using Regression Linear Regression using Scikit-Learn Principal Component Analysis Outlier Detection Using K-Means Clustering Chapter 9: Case Studies Case Study 1: Human Emotion Classification Methodology Dataset Interfacing the Raspberry Pi with MindWave Mobile via Bluetooth Data Collection Process Features Taken from the Brain Wave Signal Unstructured Data to Structured Dataset Exploratory Data Analysis from the EEG Data Classifying the Emotion Using Learning Models Case Study 2: Data Science for Image Data Exploratory Image Data Analysis Preparing the Image Data for Model Object Detection Using a Deep Neural Network Case Study 3: Industry 4.0 Raspberry Pi as a Localized Cloud for Industry 4.0 Collecting Data from Sensors Preparing the Industry Data in the Raspberry Pi Exploratory Data Analysis for the Real-Time Sensor Data Visualizing the Real-Time Sensor Data Report Generation by Reading Bar Codes Using Vision Cameras Transmitting Files or Data from the Raspberry Pi to the Computer References Index