DALI Release 0.14.0 Beta

Using DALI 0.14.0 Beta

The DALI 0.14.0 can be used with the 19.09 NVIDIA GPU Cloud (NGC) optimized container for MXNet, PyTorch, and TensorFlow.

To upgrade to DALI 0.14.0 beta from an older version of DALI, follow the installation instructions in the DALI Quick Start Guide.

Refer to the DALI Developer Guide for usage details.

Note: The internal DALI C++ API used for operators implementation, and the C++ API that enables using DALI as a library from native code, are not yet officially supported. Hence these APIs may change in the next release without advance notice.

Key Features and Enhancements

This DALI release includes the following key features and enhancements.

  • Added an example for using DALI with multiple GPUs.
  • Added a Shapes operator which returns the input shape as a tensor.
  • Added conda recipes for DALI used for pre-built packages in Watson Machine Learning Community Edition (IBM Power and x86 systems).
  • Extended the ExternalSource operator so it can hold more than one batch of inputs at a time.

Fixed Issues

This DALI release includes the following fixes.

  • Fixed FP16 bug from #1129 and added FP16 test case. (#1160)
  • Fixed framework iterators behavior when iter_setup raises StopIteration. (#1136)
  • Fixed nvjpeg legacy API. (#1179)
  • Fixed conversions to INT64 and UINT64. (#1205)
  • Fixed DALI TensorFlow install for conda environments. (#1214)

Breaking Changes

  • Extended external source operator capacity (#1127) - it now requires input to be set for every iteration.
  • Adjusted Operator::Run to take reference instead of pointer (#1168) (C++ Backend API).

Deprecated Features

  • Python .cpu() function for EdgeReference was removed to reduce confusion. (#1181)

Known Issues

  • The new video reader operator requires NVIDIA VIDEO CODEC SDK support in the platform. Prior to 19.01, the NVIDIA GPU Cloud (NGC) optimized containers lack this functionality in the default configuration. To enable the functionality, run the container with the "video" capability enabled, as shown below: 
    -e "NVIDIA_DRIVER_CAPABILITIES=compute,utility,video"
  • The video loader operator requires that the key frames occur at a minimum every 10 to 15 frames of the video stream. If the key frames occur at a lesser frequency, then the returned frames may be out of sync.
  • DALI TensorFlow plugin may be not compatible with TensorFlow 1.14.0 release. The DALI TensorFlow plugin requires that the gcc compiler that matches the one used to build TensorFlow (gcc 4.8.4 or gcc 4.8.5, depending on the particular version) be present on the system.