tensor
n. countablen. a mathematical object that describes relationships between other mathematical objects, like numbers or vectors. You might hear this word when people talk about how computers learn to recognize images or speech.
n. a mathematical object that generalizes the concepts of scalars and vectors to higher dimensions. It represents a multi-linear map from a set of vectors to a scalar, frequently utilized in physics for describing stress or in machine learning for multi-dimensional data arrays.
The software uses a tensor to represent the image data.
In general relativity, a tensor is used to describe how gravity curves the fabric of space and time.
Modern deep learning frameworks are built around the efficient manipulation of tensors, allowing the system to process massive amounts of high-dimensional data simultaneously across multiple layers.
Borrowed from New Latin tensor (“that which stretches”), equivalent to tense + -or. Anatomical sense from 1704. Introduced in the 1840s by William Rowan Hamilton as an algebraic quantity unrelated to the modern notion of tensor. The contemporary mathematical meaning was introduced (as German Tensor) by Woldemar Voigt (1898) and adopted in English from 1915 (in the context of general relativity), obscuring the earlier Hamiltonian sense. The mathematical object is so named because an early application of tensors was the study of materials stretching under tension. (See, for example, Cauchy stress tensor on Wikipedia.Wikipedia)
Commonly used in the plural form 'tensors' when referring to the data structures in computational contexts.