Classification
Classification is the task of assigning an input or observation to one or more categories from a defined set of labels.
Vibe Engineering
Classification is a supervised machine-learning technique that assigns inputs to one of a defined set of classes.
- How it is used
- Assigns medicines to High Risk or Low Risk classes using text features and interaction-derived labels.
- How deep it goes
- Traces tokenization, padding, embeddings, dropout, GRU processing, and sigmoid output.
- Examples
- Uses medicine descriptions, side effects, interaction keywords, a generated notebook, and a confusion matrix.
- What you can do after
- Evaluate a drug-risk classifier by checking class counts, training behavior, and high-risk precision and recall.
Building LLM Applications with DSPy
The task of assigning an input to one or more predefined categories.
- How it is used
- Assigns customer messages to predefined intents, including an Unknown option when the intent cannot be determined.
- How deep it goes
- Distinguishes class count, labels per item, and fixed or evolving classes, then outlines a DSPy baseline without detailing classification algorithms.
- Examples
- Uses customer-intent labels such as Update Email and Unknown, plus a multilabel flight request involving travel and food options.
- What you can do after
- Build a baseline LM classifier by specifying a model, DSPy signature, and module, then inspect prompts and usage.
Read how Building LLM Applications with DSPy explains classification →
Also covered in
- Transfer Learning for Natural Language Processing
- Timeless Algorithms, The Foundational IdeasIn early access (MEAP)
- AI Model EvaluationIn early access (MEAP)
- Grokking Statistics
What to know first
- machine learningMachine learning is a field of computational methods in which systems learn patterns or relationships from data to make predictions, decisions, or other outputs.
- precisionPrecision is the proportion of items predicted, selected, or retrieved as positive that are correct or relevant.
- recallRecall is the proportion of all relevant or actual positive items that a system successfully retrieves or correctly identifies.
- reliabilityReliability is the degree to which a system, process, or measurement performs or yields consistent, dependable results under specified conditions.
Related concepts
- query-complexity classificationincludes · Query-complexity classification is the assignment of queries to categories based on the difficulty or reasoning capability required to answer them.
- [CLS] tokenused by · A [CLS] token is a designated special token in a Transformer input whose contextual representation can aggregate information for sequence- or image-level classification.
- cross entropy lossused by · Cross-entropy loss measures the expected negative log probability that a predicted distribution assigns to outcomes drawn from a target distribution.
- ELMoused by · ELMo (Embeddings from Language Models) is a deep contextualized word-representation model that generates context-dependent embeddings from a bidirectional language model.
Questions about classification
What is classification?
Classification is the task of assigning an input or observation to one or more categories from a defined set of labels.
Which Manning books explain classification?
Vibe Engineering and Building LLM Applications with DSPy.
What should I know before learning classification?
Manning books that teach classification usually introduce machine learning, precision, recall and reliability first.
What kinds of classification are there?
Kinds of classification include birth sex classification, query-complexity classification and binary classification.
What is classification also called?
Classification is also called classify.
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.

