Verified project record

tensorflow/tensorflow

An end-to-end open source machine learning platform providing a comprehensive, flexible ecosystem of tools, libraries, and community resources for research and application development. Users can build and train models using stable Python and C++ APIs across Linux, macOS, Windows, Android, and Raspberry Pi.

Project overview

Provides a complete platform for conducting machine learning research and building ML-powered applications, supported by stable Python and C++ APIs and cross-platform official builds.

Project type: Infrastructure
Deployment: Refer to project documentation
License: Apache-2.0

Best for

Researchers and developers who need a flexible, comprehensive ecosystem with stable Python and C++ APIs to build and train machine learning models.
Educators and learners seeking a framework with broad community resources and official builds for standard desktop and single-board computer environments.

Key capabilities

Provides the core framework, including tools and libraries, required to build and train machine learning models.
Offers stable Python and C++ APIs. APIs for other languages are available but do not guarantee backward compatibility.
Includes support for CUDA-enabled GPU cards on Ubuntu and Windows, as well as DirectX and MacOS-metal via Device Plugins.
Distributes official builds for Linux (CPU/GPU/XLA), macOS, Windows (CPU/GPU), Android, and Raspberry Pi.

Limitations and risks

While Python and C++ APIs are stable, APIs for other programming languages do not guarantee backward compatibility.
GPU acceleration is limited to specific configurations, including CUDA-enabled cards on Ubuntu and Windows, DirectX, and MacOS-metal.
Telemetry practices are not documented in the available facts.

Getting started

Installation is rated as easy and can be accomplished using the command: pip install tensorflow. The framework also supports multi-route installation guides.
Users must write Python or C++ code to interact with the framework as a library to perform machine learning operations.

Evidence and sources

Open original GitHub project