Beginner · Data
Dense feature
Visual diagram · (in preparation) · Math · (in preparation) · Worked example · 3 difficulty levels.
TL;DR. A feature where most values are non-zero, typically represented as a floating-point tensor.
Technical Definition
A feature where most values are non-zero, typically represented as a floating-point tensor.
How it works
A dense feature is a type of data representation where the majority of values are non-zero. This is common in machine learning, especially when features are represented as dense vectors or tensors, such as embeddings. Dense features are generally easier for models to process directly compared to sparse features.
Related Concepts
- Feature vector — A feature vector is a numerical representation of an example, composed of its feature values, used as input for machine learning models.
- Tensor — A multidimensional array used to represent data in many dimensions, fundamental to deep learning.