Ask or search…



In this glossary, you will find a comprehensive list of terms related to Edge Impulse and various related fields. The terms are organized alphabetically within their respective categories for easy reference. Explore the sections:

Edge Impulse Specific Terms

Platform Features

  • Data pipeline: In Edge Impulse a data pipeline can be part of a project or be stand-alone. In the case in which a data pipeline is included in a project it can be composed of multiple data processing steps all connected sequentially, with the last step responsible for importing data into a single project. Stand-alone data pipelines will be counted as additional projects.
  • Edge Impulse Studio: Development platform for AI on edge devices.
  • EON (Edge Optimized Neural) Tuner: Auto ML tool for finding the best combination of DSP and ML models in Edge Impulse.
  • FOMO (Faster Objects More Objects): Technique for object detection in Edge Impulse.
  • Impulse: An on-device optimized processing pipeline composed of a combination of preprocessing, DSP and ML models.
  • Ingestion Service: Collecting and transferring data to Edge Impulse.
  • ML pipeline: A machine learning pipeline is the set of processes responsible for the flow of data into, and out from a machine learning model (or set of multiple models). It includes a dataset, processed features, a set of one or more DSP and machine learning models, and prediction outputs.
  • Private Project: A private project is an Edge Impulse project only viewable, modifiable and clonable by the user, collaborators and organization members.
  • Project: An Edge Impulse project is an ML pipeline. Read Enterprise ToS for full definition.
  • Public Project: A public project is an Edge Impulse project made public under the Apache 2.0 license with the community on the Edge Impulse community portal. This project is viewable, inspectable and cloneable by anyone on the internet.
  • User: Designated license holder on the Edge Impulse platform, empowered to create and execute projects.

Edge AI Terms

  • Microcontroller (MCU): A small computer on a single integrated circuit, often used in IoT devices.
  • Neural Processing Unit (NPU): Specialized hardware for efficient neural network computations.
  • Real-Time Operating System (RTOS): An OS designed for real-time applications.

Machine Learning (ML) Terms

  • Algorithm: A procedure used for solving a problem or performing a computation. An ML model is an example of an algorithm.
  • Quantization: A technique to reduce model weights and biases in numerical precision to save memory and speed up computation, sometimes at the cost of model accuracy.
  • Float32: A type of numerical precision where each number is stored with decimal values and take 32 bits of computer memory to store.
  • Int8: A type of numerical precision where each number is stored as a whole number and take 8 bits of computer memory to store.
  • Artificial Intelligence (AI): The simulation of human intelligence in machines.
  • Data Preprocessing: Cleaning and organizing raw data before model training.
  • Deep Learning: ML subset using neural networks with many layers.
  • Machine Learning (ML): AI field enabling systems to learn and improve from experience.
  • Training: Teaching a machine learning model using data.
  • Inference: Making predictions using a trained ML model.

IoT (Internet of Things) Terms

  • Connectivity: Methods and technologies for connecting IoT devices.
  • Firmware: Software programmed into the read-only memory of electronic devices.
  • IoT Device: A device connected to the Internet with computing capabilities.
  • Sensor Data: Data from physical sensors like temperature, motion, etc.

Industrial IoT (IIoT) Terms

  • Condition Monitoring: Tracking machine or component health.
  • Digital Twin: A virtual model of a physical process or product.
  • Edge Computing: Processing data near its generation point.
  • Industrial Automation: Control systems for industrial process management.
  • Machine-to-Machine (M2M): Direct communication between devices.
  • PLC: Controller for industrial processes.
  • Predictive Maintenance: Using IoT for maintenance predictions.
  • Prescriptive Maintenance: Maintenance strategy that uses data analysis and diagnostics.
  • SCADA: System for remote monitoring and control.
  • Sensor Fusion: Combining data from multiple sensors for accuracy.

Health and Medical Terms

  • Bioinformatics: Computational technology in molecular biology.
  • Biometric Monitoring: Tracking physiological data for health purposes.
  • Medical Imaging: Visual representations of the interior of a body.
  • Personal Health Records (PHRs): Health records maintained by the patient.
  • Remote Patient Monitoring (RPM): Recording and analyzing health data in real-time.
  • Smart Health: Advanced technologies in healthcare for monitoring and treatment.
  • Telehealth: Health-related services via electronic technologies.
  • Wearable Technology: Devices collecting health and exercise data.

TinyML Terms

  • Inference: Making predictions using a trained ML model.
  • Model Compression: Reducing machine learning model size and complexity.
  • Tiny Machine Learning (TinyML): ML for low-power devices.
  • Model: A combination of algorithm and state that is trained to perform a particular task.
  • Dataset: A collection of data samples gathered in order to train a model.
  • Metadata: Additional information that is associated with a given data sample in the dataset.
  • Label: A special type of metadata that is used during training to instruct the model on some property of the data.
  • Training: The computational process that modifies a model’s internal state based on a dataset, allowing it to perform a task.
  • Classification: The task of determining what category (or class) an input belongs to.
  • Regression: The task of predicting a numeric value based on an input.
  • Object detection: The task of identifying and localizing specific objects within an image.
  • Transfer learning: A special technique for training models that reduces the amount of data required.
  • Neural network: A model whose structure is inspired by networks of biological neurons.
  • Deep learning: An approach to creating and training models that are based on stacked layers of neural networks.
  • Classical ML: The broader category of ML models that are not based on neural networks.
  • TensorFlow: A set of software tools focused on deep learning, published by Google.
  • Keras: A tool within TensorFlow that makes it easy to create and train deep learning models.
  • TensorFlow Lite: A tool within TensorFlow that helps run inference on mobile and embedded Linux devices.
  • TensorFlow Lite for Microcontrollers: A tool within TensorFlow that helps run inference on bare metal devices such as microcontrollers.

Microcontroller Units (MCUs) Terms

  • ADC (Analog-to-Digital Converter): Converts an analog signal into digital.
  • Flash Memory: Non-volatile memory that can be electronically erased and reprogrammed.
  • GPIO (General-Purpose Input/Output): Pins on an MCU controlled by the user.
  • Interrupt: Signal indicating a need for immediate attention.
  • I2C (Inter-Integrated Circuit): Communication bus for connecting peripheral ICs to processors.
  • MCU (Microcontroller Unit): A compact integrated circuit for specific operations.
  • PWM (Pulse Width Modulation): Technique for analog results with digital means.
  • UART (Universal Asynchronous Receiver/Transmitter): Serial communication protocol.

Single Board Computers (SBCs) Terms

  • ARM Processor: CPUs based on the RISC architecture, used in SBCs.
  • Embedded Linux: Linux OS/kernel used in embedded systems.
  • Ethernet Port: Networking port on some SBCs.
  • GPIO Header: Group of pins on an SBC for interfacing with other circuits.
  • Heat Sink: Component for dissipating heat in SBCs.
  • Linux OS: Operating system for many SBCs, known for its open-source nature.
  • On-board Storage: Built-in storage capacity in an SBC.
  • Raspberry Pi: A series of small SBCs for teaching computer science.
  • SBC (Single Board Computer): A complete computer on one circuit board.

Embedded Systems Terms

  • Actuator: Device converting energy into motion in embedded systems.
  • Cross-Compilation: Compiling code for an embedded system with a different architecture.
  • Embedded Programming: Writing software for embedded systems.
  • Embedded System: Computer hardware and software for specific functions.
  • Real-Time Processing: Immediate processing of input for timely output.
  • Sensor Integration: Incorporating sensors into an embedded system.
  • SoC (System on Chip): An integrated circuit integrating all components of a computer.

Closing note on our glossary

The terms in this glossary are defined based on their usage in Edge Impulse documentation and tutorials. Some terms may have different meanings in other contexts. For example, the term "project" is used in Edge Impulse to refer to a machine learning pipeline, but it may have other meanings in other contexts. If you are unsure about the meaning of a term, please refer to the context in which it is used in Edge Impulse documentation.
Let us know if you have any questions or suggestions for this glossary. We are always aiming to keep up with the latest terminology in our documentation and resources please feel free to note any you feel we missed any or want to discuss these on our forum.