Skip to content

Glossary

A-M

Artificial Intelligence:

Back-End: The server-side component of a system that handles data processing, logic, and communication with databases or models. In MLOps, the back-end manages tasks such as model serving, data handling, and communicating with front-end applications.

AWS SageMaker: A fully managed cloud service provided by Amazon Web Services (AWS) that enables developers and data scientists to build, train, deploy, and monitor machine learning models at scale. It provides integrated tools for the entire ML lifecycle.

C-Stack:

Cloud Computing:

Deployment: The process of making a machine learning model available in a live environment so it can serve predictions or interact with end-users in real-time or batch processes. This usually involves setting up infrastructure to support the model.

Endpoint: A specific URL or network location where a deployed machine learning model or service is hosted, allowing external applications or users to make requests and receive responses.

Front-End: The part of a software system that users interact with directly, usually a Graphical User Interface (GUI). In machine learning applications, this could be a web interface or application that collects inputs and displays results from the machine learning model. (TODO: maybe link to the C-Stack tutorial here?)

Ground Truth: The labels, outputs, or annotations that are used to train and validate machine learning models. Model predictions are compared against the ground truth during training and evaluation to determine model performance.

Machine Learning:

N-Z

Productionise: Productionising refers to the process of taking a machine learning model from a prototype or experimental phase and deploying it in a production environment where it can be used for real-world tasks. This typically involves optimizing the model for scalability, reliability, and performance.

Prototype: An initial version of a machine learning model or system built to validate its feasibility. Often used interchangeably with Proof-of-Concept (POC).

S3 Bucket: Also known as Amazon Simple Storage Service, an S3 bucket is a scalable object storage service provided by AWS, commonly used to store datasets, model artifacts, and other resources required for machine learning workflows.

Streamlit: A popular open-source Python library that makes it easy to build and share custom web applications. It allows for rapid prototyping of interactive dashboards with minimal code.