Core classes¶
These pages explain the main classes and core concepts you need to understand how to use metatensor, without tying it to how a specific library uses metatensor to store data.
🔬 Core classes overview
Explore the core types of metatensor: TensorMap
, TensorBlock
and
Labels
, and how they are used to store data, metadata and gradients
of the data together.
💡 Tutorials
Learn how to use the core classes with Python code examples, and improve your understanding of data and metadata storage with metatensor.
Metatensor main classes are available from multiple languages, all documented below.