TensorFlow is an open source software library for high performance numerical computation. Its flexible architecture allows easy deployment of computation across a variety of platforms (CPUs, GPUs, TPUs), and from desktops to clusters of servers to mobile and edge devices.
TensorFlow was originally developed by researchers and engineers from the Google Brain team within Google’s AI organization.
|1||v2.7 for Python 2||v3.3+ for Python 3|
|2||NVIDIA CUDA 7.5||NVIDIA CUDA 8.0 required for Pascal GPUs|
|3||Hardware: NVIDIA® GPU card with CUDA® Compute Capability 3.5 or higher|
|4||Software: NVIDIA® GPU drivers —CUDA 9.0 requires 384.x or higher, CUDA® Toolkit —TensorFlow supports CUDA 9.0, CUPTI ships with the CUDA Toolkit, cuDNN SDK (>= 7.2)||Minimum Software plus: NCCL 2.2 for multiple GPU support, TensorRT 4.0 to improve latency and throughput for inference on some models|
This page was last updated with commit: Fixed syntax in software source files (1e763e9)