Single-label categorical classification
A classification setup in which each input is assigned no more than one label from a set of classes.
Deep Learning with Python, Third Edition
A classification task in which each example belongs to exactly one class from a set of mutually exclusive classes.
- How it is used
- Configures neural-network outputs for examples that belong to exactly one mutually exclusive class.
- How deep it goes
- Specifies output width, activation, and target-dependent loss, but stops short of explaining training or probability computation.
- Examples
- Uses a Keras dense network with two hidden layers and a softmax output.
- What you can do after
- Build it from scratch with Keras by matching output units, activation, loss, and target encoding.
Read how Deep Learning with Python, Third Edition explains single-label categorical classification →
Building LLM Applications with DSPy
A single-label problem is a classification setup in which each input is assigned at most one class label.
- How it is used
- Frames user-intent classification around one intent per utterance and simplifies ATIS labels by removing combined-intent classes.
- How deep it goes
- Defines label cardinality, gives a multi-intent counterexample, and notes that the simpler setup can be too limiting.
- Examples
- Uses a flight-and-meals request as a multi-intent counterexample and removes combined-intent classes from ATIS.
- What you can do after
- Apply the single-label assumption by excluding combined-intent classes when preparing an ATIS intent classifier.
Read how Building LLM Applications with DSPy explains single-label categorical classification →
Related concepts
- multilabel classificationcontrasts with · Multilabel classification is a classification task in which each input may be assigned zero or more labels simultaneously.
- classificationkind of · Classification is the task of assigning an input or observation to one or more categories from a defined set of labels.
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.