Product & software
TensorFlow
TensorFlow is a platform for building and running machine-learning models. Its tools cover tensor operations, model training, and deployment workflows, with documentation for both introductory use and more specialized applications.
At a glance
- Primary focus
- Machine-learning development
- Category
- machine learning
Overview
TensorFlow is a platform for building and running machine-learning models. Its tools cover tensor operations, model training, and deployment workflows, with documentation for both introductory use and more specialized applications.
How it is used
Developers define a model, prepare training data, and evaluate the resulting predictions. TensorFlow's ecosystem provides higher-level model-building interfaces as well as lower-level operations.
A useful distinction
Training a model and serving predictions involve different constraints. An evaluation should consider the target environment, input data, and model behavior rather than framework selection alone.
Sources and review
MOOR's explanatory text is supported by the following source links.
- TensorFlow: official overview — TensorFlow
- TensorFlow: documentation and further reading — TensorFlow