Releases: mind/wheels
TensorFlow 1.9 (CPU)
TensorFlow 1.9 (GPU, CUDA 9.1, cuDNN 7.1, no MKL)
GPU wheels for TensorFlow 1.9, built by TinyMind, the cloud machine learning platform.
To use the wheels on your own machine, Intel Broadwell or later CPU, and Nvidia computing capability 3.7 or later GPU with CUDA 9.1 and cuDNN 7.1 are required.
This version is optimized for compute capabilities 3.7 (K80, AWS P2/GCP), 6.0 (P100, GCP) and 7.0 (V100, AWS P3).
TensorFlow 1.8 (GPU, CUDA 9, TensorRT, no MKL)
GPU wheels for TensorFlow 1.8, built by TinyMind, the cloud machine learning platform.
To use the wheels on your own machine, Intel Broadwell or later CPU, and Nvidia computing capability 3.7 or later GPU with CUDA 9 and CuDNN 7 are required.
This version is optimized for compute capabilities 3.7 (K80, AWS P2/GCP), 6.0 (P100, GCP) and 7.0 (V100, AWS P3). TensorRT 3.0 is supported.
TensorFlow 1.8 (GPU, CUDA 9.1, cuDNN 7.1, no MKL)
GPU wheels for TensorFlow 1.8, built by TinyMind, the cloud machine learning platform.
To use the wheels on your own machine, Intel Broadwell or later CPU, and Nvidia computing capability 3.7 or later GPU with CUDA 9.1 and cuDNN 7.1 are required.
This version is optimized for compute capabilities 3.7 (K80, AWS P2/GCP), 6.0 (P100, GCP) and 7.0 (V100, AWS P3).
TensorFlow 1.8 (CPU)
CPU wheels for TensorFlow 1.8, built by TinyMind, the cloud machine learning platform.
To use the wheels on your own machine, Intel Broadwell or later CPU is required.
TensorFlow 1.14 (CPU, MKL)
CPU wheels for TensorFlow 1.14, built by TinyMind, the cloud machine learning platform. To use the wheels on your own machine, Intel Broadwell or later CPU is required.
We use Compilation flags --config=mkl --copt=-mavx2 --copt=mavx512f --copt=-mfma in our Broadwell CPU machine. Built for Ubuntu 18.04 LTS .
NOTE: These wheels contain MKL support. If you don't have it, install MKL by following the instructions here.
TensorFlow 1.13 (GPU, CUDA 10.0, cuDNN 7.4, TensorRT, no MKL)
GPU wheels for TensorFlow 1.13, built by TinyMind, the cloud machine learning platform.
To use the wheels on your own machine, Intel Broadwell or later CPU, and Nvidia computing capability 3.7 or later GPU with CUDA 10.0 and cuDNN 7.4 are required.
We use Compilation flags --copt=-mavx in our Broadwell CPU machine. Built for Ubuntu 18.04 LTS .
This version is optimized for compute capabilities 3.7 (K80, AWS P2/GCP), 6.0 (P100, GCP), 6.1(GTX1050/1060/1080/1080Ti) and 7.0 (V100, AWS P3).
TensorRT 5.0 is supported.
TensorFlow 1.13 (CPU, MKL)
CPU wheels for TensorFlow 1.13, built by TinyMind, the cloud machine learning platform. To use the wheels on your own machine, Intel Broadwell or later CPU is required.
We use Compilation flags --config=mkl --copt=-mavx2 --copt=-mfma in our Broadwell CPU machine. Built for Ubuntu 18.04 LTS .
NOTE: These wheels contain MKL support. If you don't have it, install MKL by following the instructions here.
TensorFlow 1.12 (GPU, CUDA 10.0, cuDNN 7.4, TensorRT, no MKL)
GPU wheels for TensorFlow 1.12, built by TinyMind, the cloud machine learning platform.
To use the wheels on your own machine, Intel Broadwell or later CPU, and Nvidia computing capability 3.7 or later GPU with CUDA 10.0 and cuDNN 7.4 are required.
We use Compilation flags -march=native in our Broadwell CPU machine. Built for Ubuntu 18.04 LTS .
This version is optimized for compute capabilities 3.7 (K80, AWS P2/GCP), 6.0 (P100, GCP), 6.1(GTX1050/1060/1080/1080Ti) and 7.0 (V100, AWS P3).
TensorRT 5.0 is supported.
TensorFlow 1.12 (GPU, CUDA 10.0, cuDNN 7.4, no TensorRT, no MKL)
GPU wheels for TensorFlow 1.12, built by TinyMind, the cloud machine learning platform.
To use the wheels on your own machine, Intel Broadwell or later CPU, and Nvidia computing capability 3.7 or later GPU with CUDA 10.0 and cuDNN 7.4 are required.
We use Compilation flags -march=native in our Broadwell CPU machine. Built for Ubuntu 18.04 LTS .
This version is optimized for compute capabilities 3.7 (K80, AWS P2/GCP), 6.0 (P100, GCP), 6.1(GTX1050/1060/1080/1080Ti) and 7.0 (V100, AWS P3).
TensorRT is not supported.