PyTorch Release 17.05
The NVIDIA container image of PyTorch, release 17.05, is available.
Contents of PyTorch
This container image contains the complete source of the version of PyTorch in
/opt/pytorch. It is pre-built and installed into the
/usr/local/[bin,share,lib] directories in the container image.
The container also includes the following:
- Ubuntu 16.04
- NVIDIA CUDA® 8.0.61
- NVIDIA CUDA® Deep Neural Network library™ (cuDNN) 6.0.21
- NVIDIA® Collective Communications Library ™ (NCCL) 1.6.1 (optimized for NVLink™ )
Key Features and Enhancements
- Latest cuDNN release
- Ubuntu 16.04 with April 2017 updates
The NCCL library version 1.6.1 included in this image, modifies the output buffers on all GPUs during in-place ncclReduce() operations, whereas normally only the "root" (target) device's output buffer should be modified. This is fixed in later versions of NCCL, as will be packaged in later versions of this image. As a workaround, either use ncclAllReduce(), which correctly modifies output buffers of all GPUs to the same values, or use out-of-place ncclReduce(), wherein the output buffer is distinct from the input buffer.