Release Notes

This document describes the key features, software enhancements and improvements, and known issues for DALI 1.33.0. For previously released DALI documentation, see DALI Archives.

Overview

DALI offers both performance and flexibility of accelerating different data pipelines (graphs that can have multiple outputs and inputs), as a single library, that can be easily integrated into different deep learning training and inference applications.

Using DALI

Note

DALI builds for NVIDIA® CUDA® 12 dynamically link the CUDA toolkit. To use DALI, install the latest CUDA toolkit.

To upgrade to DALI 1.33.0 from a previous version of DALI, follow the installation and usage information in the DALI User Guide.

Note

The internal DALI C++ API used for operator’s implementation, and the C++ API that enables using DALI as a library from native code, is 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:

  • Enhanced experimental support for checkpointing (saving and resuming DALI pipelines at arbitrary iteration) (5232, 5195, and 5166).

  • Improved Python annotations and signatures (5217, 5159, 5167,5154, 5188, 5158, and 5150).

    • Added annotations for JAX and Pytorch iterators (5129 and 5197).

    • Improved PythonFunction annotations (5207 and 5149).

    • Improved data type annotations (5179 and 5153).

  • Improved JAX support:

    • Added pmap compatibility for the JAX data_iterator (5185).

    • Improved JAX, Flax, and Paxml training examples (5176 and 5205).

  • Moved to CUDA 12.3U1 and enabled NVIDIA GPUDirect Storage and nvJPEG2k support for the SBSA platform (5209 and 5170).

  • Added Python3.11 support and experimental support for Python 3.12 (5174).

Fixed Issues

The following fixes are included in this release:

  • Fixed the fn.normalize handling of batch of empty samples (5223).

  • Fixed the infinite video decoder seek loop (5218).

  • Fixed the computation of maximal threads number for kernels in GPU fn.transpose and fn.normalize (5208).

  • Fixed the handling of empty slices and slicing of empty inputs (5204).

  • Fixed the scalar constant dimensionality inference (5191).

  • Fixed the sharding in the Caffe reader (5172).

Breaking Changes

There are no breaking changes in this release.

Deprecated Features

There are no deprecated features in this release.

Known Issues

This DALI release includes the following known issues:

  • The experimental.readers.fits, experimental.decoders.video, experimental.inputs.video, random_resized_crop, and experimental.decoders.image_random_crop operators do not currently support checkpointing.

  • 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 frequency that is less than 10-15 frames, the returned frames might be out of sync.

  • The experimental VideoReaderDecoder does not support open GOP.

    It will not report an error and might produce invalid frames. VideoReader uses a heuristic approach to detect open GOP and should work in most common cases.

  • The DALI TensorFlow plug-in might not be compatible with TensorFlow versions 1.15.0 and later.

    To use DALI with the TensorFlow version that does not have the prebuilt plug-in binary that is shipped with DALI, ensure that the compiler that is used to build TensorFlow exists on the system during the plug-in installation. (Depending on the particular version, you can use GCC 4.8.4, GCC 4.8.5, or GCC 5.4.)

  • In experimental debug and eager modes, the GPU external source is not properly synchronized with DALI internal streams.

    As a workaround, you can manually synchronize the device before returning the data from the callback.

  • Due to some known issues with meltdown/spectra mitigations and DALI, DALI shows the best performance when running in Docker with escalated privileges, for example:

    • privileged=yes in Extra Settings for AWS data points

    • --privileged or --security-opt seccomp=unconfined for bare Docker