Scale
Scale is the size or magnitude of a model, system, dataset, workload, or process, often in relation to its resource demands or capacity.
Software Engineering for Data Scientists
Scale is the size of a data or computational problem, commonly characterized by its number of observations, columns, memory requirements, or efficiency demands.
- How it is used
- Frames how machine-learning workflows can serve more users and process larger datasets.
- How deep it goes
- Defines scale through memory and efficiency, naming parallelization and data structures while deferring detailed approaches.
- Examples
- Uses a model serving millions of users and growing, diverse machine-learning datasets as motivating cases.
- What you can do after
- Evaluate/understand when an ML workflow needs engineering for memory and efficiency demands.
Read how Software Engineering for Data Scientists explains scale →
Machines that Think
The magnitude of a model and its associated data and computational resources.
- How it is used
- Frames large language model capabilities and limitations through the resources required to train and deploy them.
- How deep it goes
- Explains training, deployment, research, and participation constraints at an overview level.
- Examples
- Contrasts testing ideas on smaller models with evaluation at frontier scale and identifies GPUs and TPUs as specialized hardware.
- What you can do after
- Evaluate how resource requirements affect model usability, research choices, and participation in development.
Also covered in
- Graph Neural Networks in Action
- Understanding AIIn early access (MEAP)
- Hidden InfluencesIn early access (MEAP)
What to know first
- trainingTraining is the process of fitting a machine-learning model to data by adjusting its parameters or learned representations to improve its performance on a task.
- language modelA language model is a computational model that represents patterns in linguistic sequences to assign probabilities, predict, generate, or evaluate language.
Related concepts
- costconfigured by · Cost is the monetary expenditure or other resource burden incurred to acquire, produce, operate, or perform something.
- cloud scaleincludes · A scale of application use characterized by a potentially very large number of users.
- error ratecontrasts with · Error rate is the proportion of evaluated cases, operations, or outputs that produce an error or fail a specified criterion.
- trainingconfigures · Training is the process of fitting a machine-learning model to data by adjusting its parameters or learned representations to improve its performance on a task.
Questions about scale
What is scale?
Scale is the size or magnitude of a model, system, dataset, workload, or process, often in relation to its resource demands or capacity.
Which Manning books explain scale?
Software Engineering for Data Scientists and Machines that Think.
What should I know before learning scale?
Manning books that teach scale usually introduce training and language model first.
What is the difference between scale and manual labeling?
Scale is the size or magnitude of a model, system, dataset, workload, or process, often in relation to its resource demands or capacity. The assignment of labels to data by human reviewers rather than by an automated process.
What kinds of scale are there?
Kinds of scale include cloud scale.
These guides are generated from the text of Manning books by Manning's knowledge graph and link to the sections they draw on. The full books are available on liveBook and at manning.com.

