Skip to content

Releases: NVIDIA/DALI

DALI v1.50.0

27 May 17:43
@stiepan stiepan
v1.50.0
d5c7b54
Compare
Choose a tag to compare
DALI v1.50.0 Latest
Latest

Key Features and Enhancements

This DALI release includes the following key features and enhancements:

  • Added support for CUDA 12.9 (#5908)
  • Added option to disable SSL verification for S3 bucket (#5907)
    Thank you @dimabasow for your contribution.
  • Added support for loading nvComp from a Python wheel (#5894, #5889, #5909)
  • Improved error messages in video loader with file name in the message (#5910)

Fixed Issues

  • Fixed handling of multiple frames per packet in video decoder (#5911)
  • Fixed sparse tensor handling in TF plugin (#5916, #5887)
  • Fixed serialization of default seeds in operators (#5919)
  • Fixed handling of empty inputs in GPU reductions (#5914)
  • Fixed handling of stdin descriptor in CUFileDriverScope (#5902)

Improvements

  • Make Python 3.10 a default version for the build.sh (#5913)
  • Make library bundling errors easier to find in the log. (#5915)
  • Move to CUDA 12.9 (#5908)
  • Improve error messages in video loader with file name in the message (#5910)
  • Add an ability to disable SSL verification for S3 bucket (#5907)
  • Migrate DALI TF plugin to C API 2.0 (#5904)
  • BLD: Use CMake nvimgcodec module if available to get headers (#5906)
  • C API changes required for TF plugin. (#5898)
  • Remove redundant imports from the augmentation_gallery (#5900)
  • Move to externally provided nvComp (#5894)
  • Remove Python 3.8 support due to EOL (#5896)
  • Extend EfficientNet readme (#5895)
  • Fix memory consumption by PyTorch in dlpack zero-copy perf test. (#5891)
  • Add handling for NVMLError_NotSupported in get_device_memory_info (#5890)
  • Enable nvComp for SBSA platform (#5889)
  • experimental video reader to drop frames with negative display timestamps (#5885)

Bug Fixes

  • Coverity check 04.2025 (#5912)
  • frames decoder fixes: avoid overflow, handle multiple frames per packet (#5911)
  • Fix sparse tensor construction in TF plugin. (#5916)
  • Do not serialize default seed (#5919)
  • Fix gpu empty reductions (#5914)
  • Make sure that nvComp is bundled also when WITH_DYNAMIC_CUDA_TOOLKIT is off (#5909)
  • Improve conda build recipe (#5905)
  • Fix stdin handling in CUFileDriverScope (#5902)
  • Remove squeezing from C API - it probably never worked anyway. (#5893)
  • Fix invalid stack read in legacy C API (#5892)
  • Use Polygon(..., closed=true) instead of Polygon(..., true) (#5842)
  • Fix handling of scalars in TF sparse tensors. (#5887)

Breaking API changes

DALI 1.49 was the last release to support Python 3.8

Deprecated features

Support for CUDA 11 will end in the upcoming releases.

Known issues:

  • The following operators: experimental.readers.fits, experimental.decoders.video, and experimental.inputs.video 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.
  • 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 plugin might not be compatible with TensorFlow versions 1.15.0 and later.
    To use DALI with the TensorFlow version that does not have a prebuilt plugin binary shipped with DALI, make sure that the compiler that is used to build TensorFlow exists on the system during the plugin 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 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.

Binary builds

NOTE: DALI builds for CUDA 12 dynamically link the CUDA toolkit. To use DALI, install the latest CUDA toolkit.

CUDA 11.0 and CUDA 12.0 builds use CUDA toolkit enhanced compatibility. 
They are built with the latest CUDA 11.x/12.x toolkit respectively but they can run on the latest, 
stable CUDA 11.0/CUDA 12.0 capable drivers (450.80 or later and 525.60 or later respectively).
However, using the most recent driver may enable additional functionality. 
More details can be found in enhanced CUDA compatibility guide.

Install via pip for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda120==1.50.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda120==1.50.0

or just:

pip install nvidia-dali-cuda120==1.50.0
pip install nvidia-dali-tf-plugin-cuda120==1.50.0

For CUDA 11:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda110==1.50.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda110==1.50.0

or just:

pip install nvidia-dali-cuda110==1.50.0
pip install nvidia-dali-tf-plugin-cuda110==1.50.0

Or use direct download links (CUDA 12.0):

Or use direct download links (CUDA 11.0):

FFmpeg source code:

  • This software uses code of FFmpeg licensed under the LGPLv2.1 and its source can be downloaded here

Libsndfile source code:

Contributors

dimabasow
Assets 2
Loading

DALI v1.49.0

29 Apr 14:58
@stiepan stiepan
v1.49.0
ccb1902
Compare
Choose a tag to compare
DALI v1.49.0

Key Features and Enhancements

This DALI release includes the following key features and enhancements:

  • Improved new (experimental) C API (#5879, #5872, #5866, #5857, #5835, #5868)
  • Added support for CUDA 12.8U1 (#5850)
  • Added CPU support to dali.fn.experimental.debayer (#5832)
    Thank you @5had3z for your contribution!
  • Added support for nvImageCodec 0.5.0 (#5854)

Fixed Issues

  • Fixed race-condition in experimental image decoder (#5856)

Improvements

  • Update VERSION to 1.49.0
  • C API 2.0 Checkpointing + unblock dali.h (#5879)
  • Temporarily disable failing test (#5882)
  • Experimental Video Reader Refactoring and API Improvements (#5839)
  • Move to LLVM 20.1.2 (#5870)
  • C API 2.0: External source info (#5872)
  • Add _zmq.cpython to the address sanitizer suppression list (#5873)
  • Set minimum CMake policy version for Horovod build (#5871)
  • Pipeline refactoring (#5866)
  • Add multi-configuration performance benchmarking (#5858)
  • Sort out Python 3.8 support (#5867)
  • Moves to manylinux_2_28 (#5608)
  • Adjust test compatibility with numpy 2.x (#5862)
  • Bump up the minimum version of CMake required by ffts (#5864)
  • Remove unnecessary global declarations and add noqa comments (#5865)
  • Add fallback for missing source info in check_batch (#5861)
  • Bump nvimagecodec requirement to 0.5.0 (#5854)
  • Skip C API2 test using Mixed ImageDecoder when it's not registered. (#5857)
  • C API 2.0 Pipeline & Pipeline Outputs (#5835)
  • Update six package version constraint (#5855)
  • Add info about GIT sha to the documentation (#5853)
  • Bump up the Black version to 25.x (#5849)
  • Bump OpenCV version in conda to 4.11 (#5851)
  • Improve sanitizer configuration and suppress false positives (#5795)
  • Add Debayer CPU based on OpenCV (#5832)
  • FramesDecoder boundary handling, video utils (#5844)
  • Move to CUDA 12.8 U1 (#5850)
  • Added warp perpective tests to other test suites. (#5847)
  • Add operator statefulness info to OpSchema (#5848)
  • Bump up support TF version to 2.18 (#5840)

Bug Fixes

  • Fixed race-condition in experimental image decoder (#5856)

Breaking API changes

There are no breaking changes in this DALI release.

Deprecated features

DALI 1.49 is the last release to support Python 3.8

Known issues:

  • The following operators: experimental.readers.fits, experimental.decoders.video, and experimental.inputs.video 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.
  • 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 plugin might not be compatible with TensorFlow versions 1.15.0 and later.
    To use DALI with the TensorFlow version that does not have a prebuilt plugin binary shipped with DALI, make sure that the compiler that is used to build TensorFlow exists on the system during the plugin 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 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.

Binary builds

NOTE: DALI builds for CUDA 12 dynamically link the CUDA toolkit. To use DALI, install the latest CUDA toolkit.

CUDA 11.0 and CUDA 12.0 builds use CUDA toolkit enhanced compatibility. 
They are built with the latest CUDA 11.x/12.x toolkit respectively but they can run on the latest, 
stable CUDA 11.0/CUDA 12.0 capable drivers (450.80 or later and 525.60 or later respectively).
However, using the most recent driver may enable additional functionality. 
More details can be found in enhanced CUDA compatibility guide.

Install via pip for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda120==1.49.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda120==1.49.0

or just:

pip install nvidia-dali-cuda120==1.49.0
pip install nvidia-dali-tf-plugin-cuda120==1.49.0

For CUDA 11:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda110==1.49.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda110==1.49.0

or just:

pip install nvidia-dali-cuda110==1.49.0
pip install nvidia-dali-tf-plugin-cuda110==1.49.0

Or use direct download links (CUDA 12.0):

Or use direct download links (CUDA 11.0):

FFmpeg source code:

  • This software uses code of FFmpeg licensed under the LGPLv2.1 and its source can be downloaded here

Libsndfile source code:

Contributors

5had3z
Loading

DALI v1.48.0

25 Mar 17:57
@stiepan stiepan
v1.48.0
211abfa
Compare
Choose a tag to compare
DALI v1.48.0

Key Features and Enhancements

This DALI release includes the following key features and enhancements:

  • Improved fn.experimental.decoders.video video decoder: (#5830, #5814)
    • Improved seeking and reset behavior
    • Added support for frame padding with configurable modes
    • Added frame selection options
    • Added build_index option to control the generation of a frame index
  • Added CPU support to dali.fn.experimental.warp_perspective (#5829, #5815)
    • Thank you @5had3z for your contribution!
  • Introduced new (experimental) C API (#5796, #5797, #5798, #5799)

Fixed Issues

  • Introduced AvUniquePtr to avoid memory leaks in frames decoder (#5834)
  • Removed an unnecessary host sync in operators taking pinned inputs. (#5822)
  • Fixed host-side access to pinned CPU buffers produced with non-host order (#5820)
  • Fixed handling of empty batches in GPU arithmetic operators. (#5818)

Improvements

  • Fix data paths in TL3 short tests (#5845)
  • Revert change of batch size in SSD LT3 to 64 due to convergence problem (#5846)
  • Update VERSION to 1.48.0
  • Fix coverity issues 25/03 (#5843)
  • Bump up FFmpeg to 7.1.1 (#5838)
  • Reorganize video decoder sources (#5836)
  • Dependency update 2025-03 (#5833)
  • C API 2.0 Tensor and TensorList (#5799)
  • Update documentation of audio decoder operator (supported formats) (#5803)
  • Removes RN50 benchmark tests, move to DALI_EXTRA for RN50 DL FW iter tests (#5824)
  • Improve video decoder seeking and reset behavior (#5830)
  • Warp Perspective CPU Impl (#5829)
  • Remove ScratchpadAllocator and ScratchpadEstimator (#5810)
  • Code modernization and refactoring in Pipeline, OpSpec and InputOperator (#5826)
  • fn.experimental.decoders.video improvements (#5814)
  • C API 2.0 helpers (#5798)
  • C API 2.0 initialization and error handling (#5797)
  • Limit the max. tensor list size in TensorTest (#5823)
  • Relax DisplacementTest.Sphere constraints from 0.005 to 0.006 (#5821)
  • Restrict dm-tree version for Python 3.8 and 3.9 (#5819)
  • Add C API header and C language build test. (#5796)
  • Expose DLPack support in the docs (#5817)

Bug Fixes

  • Fix usage of unique_ptr for arrays in data_objects_test (#5837)
  • Introduce AvUniquePtr to avoid memory leaks in frames decoder (#5834)
  • tensor_shape.h warning fix (#5831)
  • Enhanced Video Codec Support and Error Handling (#5825)
  • Fix documentation for warp_perspective, requires 3x3 shape, not flattened 1D (#5815)
  • Remove an unwanted potential host sync in operators taking pinned inputs. (#5822)
  • Fix host-side access to pinned CPU buffers produced with non-host order (#5820)
  • Fix handling of empty batches in GPU arithmetic operators. (#5818)

Breaking API changes

There are no breaking changes in this DALI release.

Deprecated features

No features were deprecated in this release.

Known issues:

  • The following operators: experimental.readers.fits, experimental.decoders.video, and experimental.inputs.video 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.
  • 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 plugin might not be compatible with TensorFlow versions 1.15.0 and later.
    To use DALI with the TensorFlow version that does not have a prebuilt plugin binary shipped with DALI, make sure that the compiler that is used to build TensorFlow exists on the system during the plugin 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 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.

Binary builds

NOTE: DALI builds for CUDA 12 dynamically link the CUDA toolkit. To use DALI, install the latest CUDA toolkit.

CUDA 11.0 and CUDA 12.0 builds use CUDA toolkit enhanced compatibility. 
They are built with the latest CUDA 11.x/12.x toolkit respectively but they can run on the latest, 
stable CUDA 11.0/CUDA 12.0 capable drivers (450.80 or later and 525.60 or later respectively).
However, using the most recent driver may enable additional functionality. 
More details can be found in enhanced CUDA compatibility guide.

Install via pip for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda120==1.48.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda120==1.48.0

or just:

pip install nvidia-dali-cuda120==1.48.0
pip install nvidia-dali-tf-plugin-cuda120==1.48.0

For CUDA 11:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda110==1.48.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda110==1.48.0

or just:

pip install nvidia-dali-cuda110==1.48.0
pip install nvidia-dali-tf-plugin-cuda110==1.48.0

Or use direct download links (CUDA 12.0):

Or use direct download links (CUDA 11.0):

FFmpeg source code:

  • This software uses code of FFmpeg licensed under the LGPLv2.1 and its source can be downloaded here

Libsndfile source code:

Contributors

5had3z
Loading

DALI v1.47.0

25 Feb 18:26
@stiepan stiepan
v1.47.0
431d238
Compare
Choose a tag to compare
DALI v1.47.0

Key Features and Enhancements

This DALI release includes the following key features and enhancements:

  • Added support for DALI batched processing as a part of PyTorch DataLoader (DALI proxy):
  • Moved to JetPack 6.2 (CUDA 12.6) for Tegra builds (#5449)

Fixed Issues

  • Fixed insufficient synchronization issue in experimental image decoder (#5806)
  • Fixed memory leak in experimental video decoder (#5778)

Improvements

  • Fix CVE-2024-13176 in openssl (#5805)
  • Update VERSION to 1.47.0
  • Make frames decoder to build index without file decoding (#5809)
  • Clean up warnings (#5811)
  • Move to PyPI to download PyNvVideoCodec (#5813)
  • Dependency update 02/2025 (#5801)
  • Use DALI as default in resnet50 example (#5808)
  • Add documentation about DALI proxy in EfficientNet and ResNet examples (#5800)
  • Add acknowledgements for AWS SDK C++, curl and openssl (#5794)
  • Move to CUDA 12.8 (#5793)
  • Move to JetPack 6.2 (CUDA 12.6) (#5449)
  • Add DALI proxy option to EfficientNet example (#5791)
  • Use DALI proxy to ResNet50 example. Introduce TL3_RN50_benchmark (#5792)
  • Remove libavutils from the asan suppression list (#5783)
  • Add a typical data loading pipeline path for the EfficeintNet (#5761)
  • Remove dead executor test. (#5788)
  • Fix test_dali_proxy usage (#5784)
  • Fix TL1_decoder_perf usage of pip show (#5781)
  • Introduce (experimental) DALI proxy (#5726)
  • Move optical flow tests from specific TU test job to Ampere tests (#5771)
  • Fix minor Markdown issues in ipynb in docs (#5773)

Bug Fixes

  • Ensure allocated temp memory is usable by nvImageCodec streams, as we are skipping pre-sync due to unnecessary overhead in the general case (#5806)
  • Remove redundant nvml::Shutdown from optical flow (#5804)
  • Fix hogging memory by libaviutils (#5778)

Breaking API changes

There are no breaking changes in this DALI release.

Deprecated features

No features were deprecated in this release.

Known issues:

  • The following operators: experimental.readers.fits, experimental.decoders.video, and experimental.inputs.video 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.
  • 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 plugin might not be compatible with TensorFlow versions 1.15.0 and later.
    To use DALI with the TensorFlow version that does not have a prebuilt plugin binary shipped with DALI, make sure that the compiler that is used to build TensorFlow exists on the system during the plugin 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 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.

Binary builds

NOTE: DALI builds for CUDA 12 dynamically link the CUDA toolkit. To use DALI, install the latest CUDA toolkit.

CUDA 11.0 and CUDA 12.0 builds use CUDA toolkit enhanced compatibility. 
They are built with the latest CUDA 11.x/12.x toolkit respectively but they can run on the latest, 
stable CUDA 11.0/CUDA 12.0 capable drivers (450.80 or later and 525.60 or later respectively).
However, using the most recent driver may enable additional functionality. 
More details can be found in enhanced CUDA compatibility guide.

Install via pip for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda120==1.47.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda120==1.47.0

or just:

pip install nvidia-dali-cuda120==1.47.0
pip install nvidia-dali-tf-plugin-cuda120==1.47.0

For CUDA 11:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda110==1.47.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda110==1.47.0

or just:

pip install nvidia-dali-cuda110==1.47.0
pip install nvidia-dali-tf-plugin-cuda110==1.47.0

Or use direct download links (CUDA 12.0):

Or use direct download links (CUDA 11.0):

FFmpeg source code:

  • This software uses code of FFmpeg licensed under the LGPLv2.1 and its source can be downloaded here

Libsndfile source code:

Loading

DALI v1.46.0

31 Jan 17:44
@stiepan stiepan
v1.46.0
412ef54
Compare
Choose a tag to compare
DALI v1.46.0

Key Features and Enhancements

This DALI release includes the following key features and enhancements:

  • Added CUDA 12.8 support
  • Optimized workspace and operator specification (#5740, #5770)
  • Introduced Common Subgraph Elimination for DALI pipeline/graph (#5752, #5755)
  • Added support for nvImageCodec 0.4.1 (#5576, #5774, #5780)
  • Improved documentation for supported environment variables (#5756)
  • Made the pipeline's build call optional (#5754)

Fixed Issues

  • Fixed DALIDataType printing in global namespace (for custom C++ builds) (#5748)

Improvements

  • Disable nvimgcodec support for Xavier builds (#5780)
  • Bump nvimagecodec to 0.4.1 version in conda build (#5774)
  • Update VERSION to 1.46.0
  • Optimize OpSpec and Workspace queries (#5770)
  • Update CUTLASS to 3.6.0 (#5765)
  • Upgrade nvImageCodec to 0.4.1 version (#5576)
  • Deps update 01/2025 (#5767)
  • Change the way the CUDA public key is added to the sources keyring (#5766)
  • Add .as_cpu() to TensorCPU and TensorListCPU (#5751)
  • Make build() optional (#5754)
  • Adjust the L1 test to adhere to the normalized DALI plugin naming (#5760)
  • Add a test to verify that CSE doesn't merge external source (#5755)
  • Document environment variables. (#5756)
  • CSE - Common Subexpression (Subgraph) Elimination (#5752)
  • Add setuptools as conda build requirements (#5753)
  • Normalize wheels and sdist names to have only _ as PEP 625 asks (#5750)
  • Add usage of taskset in the TL1_decoder_perf (#5738)
  • OpSchema major rework / deprecate default seed (#5740)
  • Fix running the python-self-core-exec2 twice (#5743)

Bug Fixes

Breaking API changes

There are no breaking changes in this DALI release.

Deprecated features

  • Passing a seed argument to non-random operators is deprecated. Passing it has no effect but triggers a warning.

Known issues:

  • The following operators: experimental.readers.fits, experimental.decoders.video, and experimental.inputs.video 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.
  • 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 plugin might not be compatible with TensorFlow versions 1.15.0 and later.
    To use DALI with the TensorFlow version that does not have a prebuilt plugin binary shipped with DALI, make sure that the compiler that is used to build TensorFlow exists on the system during the plugin 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 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.

Binary builds

NOTE: DALI builds for CUDA 12 dynamically link the CUDA toolkit. To use DALI, install the latest CUDA toolkit.

CUDA 11.0 and CUDA 12.0 builds use CUDA toolkit enhanced compatibility. 
They are built with the latest CUDA 11.x/12.x toolkit respectively but they can run on the latest, 
stable CUDA 11.0/CUDA 12.0 capable drivers (450.80 or later and 525.60 or later respectively).
However, using the most recent driver may enable additional functionality. 
More details can be found in enhanced CUDA compatibility guide.

Install via pip for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda120==1.46.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda120==1.46.0

or just:

pip install nvidia-dali-cuda120==1.46.0
pip install nvidia-dali-tf-plugin-cuda120==1.46.0

For CUDA 11:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda110==1.46.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda110==1.46.0

or just:

pip install nvidia-dali-cuda110==1.46.0
pip install nvidia-dali-tf-plugin-cuda110==1.46.0

Or use direct download links (CUDA 12.0):

Or use direct download links (CUDA 11.0):

FFmpeg source code:

  • This software uses code of FFmpeg licensed under the LGPLv2.1 and its source can be downloaded here

Libsndfile source code:

Loading

DALI v1.45.0

24 Jan 15:08
@stiepan stiepan
v1.45.0
a839356
Compare
Choose a tag to compare
DALI v1.45.0

Key Features and Enhancements

This DALI release includes the following key features and enhancements:

  • Added support for CUDA 12.8 (#5711).
  • Optimized (zero-copy) transfer of outputs in JAX and PaddlePaddle plugins with dynamic executor. (#5703, #5715)

Fixed Issues

  • Fixed passing inputs transferred from GPU to CPU with .cpu() call as keyword arguments (#5732)

Improvements

  • Add CUDA 12.8 support
  • Update libjpeg2k (#5742)
  • Update VERSION to 1.45.0
  • Remove the 'build tag' from the DALI wheel name (#5736)
  • Update CV-CUDA 0.8->0.12 rapidjson (ToT), google benchmark 1.9.0->1.15.1, black 24.4.2->24.8.0 (#5733)
  • Return TensorLayout by const-reference. (#5730)
  • Factor out DALIDataType. (#5729)
  • Improve printing in hw_decoder_bench.py (#5724)
  • Replace all parameter references from double backticks to single (#5716)
  • Enable runtime and sphinx-level signatures for ops API (#5722)
  • Move to CUDA 12.6 U3 (#5719)
  • Remove the unused max_num_stream Pipeline parameter. Deprecate max_streams in Python. (#5720)
  • Remove default_cuda_stream_priority from native code and deprecate it in Python. (#5717)
  • PaddlePaddle zero copy (#5715)
  • Add handling of parameter references in Sphinx documentation (#5707)
  • JAX zero copy (#5703)
  • Use FMA in separable resampling. (#5711)
  • Use exec-dynamic in RNN-t pipeline. Minor fix to exec2. (#5706)

Bug Fixes

  • Bump the cap for numpy version in tf tests (#5741)
  • Remove TFRecordParser dependency from backend_impl (#5737)
  • Fix coverity issue (#5734)
  • Fix passing the results of .cpu() to argument inputs. (#5732)
  • Use absolute addressing for parameters (#5725)
  • Correct nvimagecodec version in conda and in installation instruction message (#5714)

Breaking API changes

There are no breaking changes in this DALI release.

Deprecated features

  • Pipeline arguments max_streams and default_cuda_stream_priority are deprecated. Passing them has no effect, but triggers a warning.

Known issues:

  • The most recent nvImageCodec (0.4.0) is currently incompatible with DALI. Python wheels starting from DALI 1.44 pin the dependency to 0.3.0, but older releases do not specify the required version explicitly. Users of previous DALI releases may need to manually install older nvImageCodec in order to use fn.experimental.decoders.image.* or, for DALI 1.39 and 1.40, fn.decoders.image.*. The compatible version can be installed with pip install nvidia-nvimgcodec-cu12~=0.3.0.
  • The following operators: experimental.readers.fits, experimental.decoders.video, and experimental.inputs.video 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.
  • 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 plugin might not be compatible with TensorFlow versions 1.15.0 and later.
    To use DALI with the TensorFlow version that does not have a prebuilt plugin binary shipped with DALI, make sure that the compiler that is used to build TensorFlow exists on the system during the plugin 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 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.

Binary builds

NOTE: DALI builds for CUDA 12 dynamically link the CUDA toolkit. To use DALI, install the latest CUDA toolkit.

CUDA 11.0 and CUDA 12.0 builds use CUDA toolkit enhanced compatibility. 
They are built with the latest CUDA 11.x/12.x toolkit respectively but they can run on the latest, 
stable CUDA 11.0/CUDA 12.0 capable drivers (450.80 or later and 525.60 or later respectively).
However, using the most recent driver may enable additional functionality. 
More details can be found in enhanced CUDA compatibility guide.

Install via pip for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda120==1.45.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda120==1.45.0

or just:

pip install nvidia-dali-cuda120==1.45.0
pip install nvidia-dali-tf-plugin-cuda120==1.45.0

For CUDA 11:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda110==1.45.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda110==1.45.0

or just:

pip install nvidia-dali-cuda110==1.45.0
pip install nvidia-dali-tf-plugin-cuda110==1.45.0

Or use direct download links (CUDA 12.0):

Or use direct download links (CUDA 11.0):

FFmpeg source code:

  • This software uses code of FFmpeg licensed under the LGPLv2.1 and its source can be downloaded here

Libsndfile source code:

Loading

DALI v1.44.0

28 Nov 18:09
@stiepan stiepan
v1.44.0
a0feb6c
Compare
Choose a tag to compare
DALI v1.44.0

Key Features and Enhancements

This DALI release includes the following key features and enhancements:

  • The dynamic executor (exec_dynamic) is no longer experimental. It supports GPU to CPU transfers and reduces memory consumption. (#5704)
  • Added support for zero-copy outputs transfer with dynamic executor. (#5684, #5673)
    • Eliminated the outputs copy in PyTorch plugin. (#5699)
  • Added dynamic executor support to TF plugin. (#5686)
  • Optimized pipeline's output contiguity handling. (#5677)

Fixed Issues

  • Restricted nvImageCodec version in DALI wheel dependencies list, as the most recent nvImageCodec (0.4.0) is incompatible. (#5709)
  • Fixed custom stream handling on non-default device in fn.external_source (#5690).
  • Fixed problem with using DALI with Python3.12 with no distutils/setuptools installed.
  • Fixed incorrect stream usage in fn.experimental.inputs.video (#5682)
  • Fixed possible hang in video decoder when rewinding near the last keyframe (#5676, #5669)
  • Fixed dont_use_mmap option handling in fn.readers.webdataset (#5683)
  • Fixed redundant usage of pinned memory in the CPU fn.readers.numpy reader (#5678)
  • Fixed dynamic executor's handling of operators that produce no outputs (#5674)

Improvements

  • Make exec_dynamic non-experimental (alternative formatting) (#5704)
  • Use zero-copy outputs with PyTorch (#5699)
  • Add Python 3.13 (experimental) support (#5692)
  • Add proper NVTX markers to Executor2. (#5694)
  • Add Efficientnet pipeline to hw_bench script (#5691)
  • Stream aware outputs (#5684)
  • Update DALI_DEPS_VERSION for new OpenSSL (#5689)
  • Add dynamic executor support to TF plugin. (#5686)
  • Make black and flake8 run independently. (#5685)
  • Update of FFmpeg to n7.1 (#5681)
  • Deps update 10 2024 (#5670)
  • Refactor operator output contiguity handling (#5677)
  • Add ready event to Tensor and TensorList. (#5673)

Bug Fixes

  • Fix nvimgcodec version check, do not install it separately in tests env (#5713)
  • Limit the upper versions of DALI wheel installation dependencies (#5710)
  • Limit the maximum version of nvimagecodec for current DALI (#5709)
  • Use exec-dynamic in RNN-t pipeline. Minor fix to exec2. (#5706)
  • Check JAX version and invoke dlpack manually for jax pre-0.4.16. (#5702)
  • Fix nose imports (#5698)
  • ExternalSource refactoring and fixing (#5690)
  • Move from deprecated distutils to packaging (#5687)
  • Make sure that the proper video stream index is used by the GPU decoder (#5682)
  • Add an ability to rewind at the end of the video (#5676)
  • Fix inverted mmap inside webdataset reader (#5683)
  • Fix the redundant usage of pinned memory in the numpy cpu reader (#5678)
  • Fix handling of tasks with zero outputs. (#5674)
  • Add an ability to retry rewind to the one before the last keyframe (#5669)

Breaking API changes

There are no breaking changes in this DALI release.

Deprecated features

No features were deprecated in this release.

Known issues:

  • The most recent nvImageCodec (0.4.0) is currently incompatible with DALI. Python wheel for DALI 1.44 pins the dependency to 0.3.0, but older releases do not specify the required version explicitly. Users of previous DALI releases may need to manually install older nvImageCodec in order to use fn.experimental.decoders.image.* or, for DALI 1.39 and 1.40, fn.decoders.image.*. The compatible version can be installed with pip install nvidia-nvimgcodec-cu12~=0.3.0.
  • The following operators: experimental.readers.fits, experimental.decoders.video, and experimental.inputs.video 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.
  • 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 plugin might not be compatible with TensorFlow versions 1.15.0 and later.
    To use DALI with the TensorFlow version that does not have a prebuilt plugin binary shipped with DALI, make sure that the compiler that is used to build TensorFlow exists on the system during the plugin 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 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.

Binary builds

NOTE: DALI builds for CUDA 12 dynamically link the CUDA toolkit. To use DALI, install the latest CUDA toolkit.

CUDA 11.0 and CUDA 12.0 builds use CUDA toolkit enhanced compatibility. 
They are built with the latest CUDA 11.x/12.x toolkit respectively but they can run on the latest, 
stable CUDA 11.0/CUDA 12.0 capable drivers (450.80 or later and 525.60 or later respectively).
However, using the most recent driver may enable additional functionality. 
More details can be found in enhanced CUDA compatibility guide.

Install via pip for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda120==1.44.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda120==1.44.0

or just:

pip install nvidia-dali-cuda120==1.44.0
pip install nvidia-dali-tf-plugin-cuda120==1.44.0

For CUDA 11:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda110==1.44.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda110==1.44.0

or just:

pip install nvidia-dali-cuda110==1.44.0
pip install nvidia-dali-tf-plugin-cuda110==1.44.0

Or use direct download links (CUDA 12.0):

Or use direct download links (CUDA 11.0):

FFmpeg source code:

  • This software uses code of FFmpeg licensed under the LGPLv2.1 and its source can be downloaded here

Libsndfile source code:

Loading

DALI v1.43.0

30 Oct 17:22
@stiepan stiepan
v1.43.0
4562f15
Compare
Choose a tag to compare
DALI v1.43.0

Key Features and Enhancements

This DALI release includes the following key features and enhancements:

  • Added DataNode methods for runtime access to batch's shape, layout, and source_info (#5650, #5648).
  • Added support for CUDA 12.6U2 (#5657)
  • Add experimental CV-CUDA resize operator (#5637)
  • Improved performance of TensorList resizing and TypeTable (#5638, #5634).
  • Improved DLPack support (to enable sharing ownership and pinned memory) (#5661).

Fixed Issues

  • Fixed cleanup of pipelines containing PythonFunction. (#5668)
  • Fixed CPU resize operator running with multiple resampling modes in a batch. (#5647)

Improvements

  • Add support for bool type for the numba operator (#5666)
  • Bump numpy version in Xavier tests. (#5663)
  • DLPack support rework (#5661)
  • Update links in DALI readme (#5660)
  • Bump required NumPy version to 1.23. (#5658)
  • Move to CUDA 12.6 update 2 (#5657)
  • Increase number of the decoder bench iterations (#5655)
  • GetProperty refactor + DataNode.property accessor (#5650)
  • Remove and forbid direct inclusion of half.hpp. (#5654)
  • Add DataNode.shape() (#5648)
  • Fix conda build for Python 3.9 (#5649)
  • Increase batch size in RN50 test for TF as on H100 it works well again (#5645)
  • Add experimental CV-CUDA resize (#5637)
  • Pin libprotobuf and protobuf to 5.24 which works with python 3.8-3.12 in conda (#5643)
  • Optimize TensorList::Resize (#5638)
  • TypeTable/TypeInfo optimization (#5634)

Bug Fixes

  • Fix Pipeline reference leak in PythonFunction. (#5668)
  • Fix constness in (Const)SampleView. Improve diagnostics. (#5664)
  • Fix issues detected by Coverity (2024.09.30) (#5652)
  • Fix CPU resize with mixed NN/other resampling filters. (#5647)
  • Fix block size in TransposeTiled kernel test. (#5641)
  • Fix the lack of the previous release in the docs switcher list (#5640)

Breaking API changes

There are no breaking changes in this DALI release.

Deprecated features

No features were deprecated in this release.

Known issues:

  • The following operators: experimental.readers.fits, experimental.decoders.video, and experimental.inputs.video 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.
  • 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 plugin might not be compatible with TensorFlow versions 1.15.0 and later.
    To use DALI with the TensorFlow version that does not have a prebuilt plugin binary shipped with DALI, make sure that the compiler that is used to build TensorFlow exists on the system during the plugin 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 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.

Binary builds

NOTE: DALI builds for CUDA 12 dynamically link the CUDA toolkit. To use DALI, install the latest CUDA toolkit.

CUDA 11.0 and CUDA 12.0 builds use CUDA toolkit enhanced compatibility. 
They are built with the latest CUDA 11.x/12.x toolkit respectively but they can run on the latest, 
stable CUDA 11.0/CUDA 12.0 capable drivers (450.80 or later and 525.60 or later respectively).
However, using the most recent driver may enable additional functionality. 
More details can be found in enhanced CUDA compatibility guide.

Install via pip for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda120==1.43.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda120==1.43.0

or just:

pip install nvidia-dali-cuda120==1.43.0
pip install nvidia-dali-tf-plugin-cuda120==1.43.0

For CUDA 11:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda110==1.43.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda110==1.43.0

or just:

pip install nvidia-dali-cuda110==1.43.0
pip install nvidia-dali-tf-plugin-cuda110==1.43.0

Or use direct download links (CUDA 12.0):

Or use direct download links (CUDA 11.0):

FFmpeg source code:

  • This software uses code of FFmpeg licensed under the LGPLv2.1 and its source can be downloaded here

Libsndfile source code:

Loading

DALI v1.42.0

30 Sep 16:53
@stiepan stiepan
v1.42.0
3c97d9b
Compare
Choose a tag to compare
DALI v1.42.0

Key Features and Enhancements

This DALI release includes the following key features and enhancements:

  • Introduced more flexible execution in the DALI pipeline with the experimental_exec_dynamic flag (#5635, #5631, #5593, #5528, #5620, #5602, #5529, #5595):
    • Enabled support for GPU-to-CPU transfers in a pipeline.
    • Added support for accessing CPU metadata of GPU outputs (e.g. shape of GPU decoded images/videos).
  • Added support for CUDA 12.6U1 (#5616).
  • Added an option to return the number of frames in the experimental video reader (#5628).

Fixed Issues

  • Fixed handling of optical flow initialization failure (#5624).

Improvements

  • Add metadata-only inputs. (#5635)
  • Schema-based input device check (#5631)
  • Enable GPU->CPU transfers (#5593)
  • Adds enable_frame_num to the experimental video reader (#5628)
  • Executor2 class implementation & tests (#5528)
  • Executor 2.0: Per-operator stream assignment policy (#5620)
  • Move to CUDA 12.6U1 (#5616)
  • Executor 2.0: Stream assignment (#5602)
  • Tasking: Test returning multiple outputs of type std::any. (#5529)
  • Patch OSS vulnerabilities (#5612)
  • Executor 2.0: Graph lowering (#5595)
  • Make DALI tests compatible with Python 3.12 (#5452)
  • Adjust the L3 perf test threshold for H100 runners (#5606)
  • Add L1 image decoder DALI test (#5601)

Bug Fixes

  • Fix multiple initialization attempts in optical flow operator. (#5624)
  • Fix null pointer access when clearing incomplete workspace payload. (#5622)

Breaking API changes

There are no breaking changes in this DALI release.

Deprecated features

No features were deprecated in this release.

Known issues:

  • The following operators: experimental.readers.fits, experimental.decoders.video, and experimental.inputs.video 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.
  • 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 plugin might not be compatible with TensorFlow versions 1.15.0 and later.
    To use DALI with the TensorFlow version that does not have a prebuilt plugin binary shipped with DALI, make sure that the compiler that is used to build TensorFlow exists on the system during the plugin 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 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.

Binary builds

NOTE: DALI builds for CUDA 12 dynamically link the CUDA toolkit. To use DALI, install the latest CUDA toolkit.

CUDA 11.0 and CUDA 12.0 builds use CUDA toolkit enhanced compatibility. 
They are built with the latest CUDA 11.x/12.x toolkit respectively but they can run on the latest, 
stable CUDA 11.0/CUDA 12.0 capable drivers (450.80 or later and 525.60 or later respectively).
However, using the most recent driver may enable additional functionality. 
More details can be found in enhanced CUDA compatibility guide.

Install via pip for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda120==1.42.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda120==1.42.0

or just:

pip install nvidia-dali-cuda120==1.42.0
pip install nvidia-dali-tf-plugin-cuda120==1.42.0

For CUDA 11:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda110==1.42.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda110==1.42.0

or just:

pip install nvidia-dali-cuda110==1.42.0
pip install nvidia-dali-tf-plugin-cuda110==1.42.0

Or use direct download links (CUDA 12.0):

Or use direct download links (CUDA 11.0):

FFmpeg source code:

  • This software uses code of FFmpeg licensed under the LGPLv2.1 and its source can be downloaded here

Libsndfile source code:

Loading

DALI v1.41.0

29 Aug 19:14
@stiepan stiepan
v1.41.0
040b354
Compare
Choose a tag to compare
DALI v1.41.0

Key Features and Enhancements

This DALI release includes the following key features and enhancements:

  • Added support for CUDA 12.6. (#5596)
  • Added fn.experimental.warp_perspective operator. (#5542, #5575)
  • Added fn.random.beta random variate sampling operator. (#5550, #5571)
  • Added fn.io.file.read operator that supports loading files from dynamically specified paths. (#5552, #5572)
  • Enabled support for more simple types in fn.python_function, fn.ones, and fn.zeros. (#5598)
  • Removed unnecessary copy of tensor arguments fed into GPU operators. (#5590)

Fixed Issues

  • Reverted the fn.decoders.image* to use legacy decoders due to performance regression in nvImageCodec. (#5582, #5578, #5586)
  • Optimized S3 downloading in TFRecord reader. (#5554)
  • Added missing validation for number of inputs in argument promotion. (#5592)
  • Added missing header to support compilation with GCC 14. (#5594)
  • Fixed empty batch handling when copying batch from cpu to gpu. (#5567)

Improvements

  • Executor 2.0: ExecGraph (#5587)
  • Enable more Python types to be supported by the DALI python function (#5598)
  • Remove usages of std::call_once. (#5599)
  • Move to CUDA 12.6 (#5596)
  • Remove MakeContiguous before CPU inputs of GPU ops. (#5590)
  • nvImageCodec related fixes (#5586)
  • Mark PropagateError as [[noreturn]] (#5589)
  • Make test_beta_distribution compatible with Python 3.8 (#5571)
  • Add default_batch_size to IterationData. (#5588)
  • Add thread_setup callback to tasking::Executor (#5581)
  • Fix librosa deprecated usage (#5579)
  • Bring back the legacy image decoder operator (#5578)
  • Extract librosa's effects.trim and stft to DALI test utils, to avoid issues with breaking changes (#5568)
  • Remove libjpeg and libtiff deps (#5569)
  • Add warp_perspective operator (#5542)
  • Remove legacy image decoder (#5559)
  • Optimize S3 downloading for TFRecord reader (#5554)
  • Add io.file.read operator (#5552)
  • Add fn.random.beta random variate (#5550)
  • Reduce the batch size in the TensorFlow RN50 L3 test (#5565)
  • Use MakeContiguous when copying CPU->CPU. (#5562)
  • Update the DALI EfficientNet example to be compatible with the latest NumPy (#5561)

Bug Fixes

  • Fixes problems with fetching LFS objects during nvImageCodec conda build (#5603)
  • Fix the --python-tag option passed to python setup.py bdist_wheel command (#5600)
  • Revert "Reintroduce "Move old ImageDecoder to legacy module and make the nvImageCodec based ImageDecoder the default" (#5470)" (#5582)
  • Adding cstdint header to support GCC 14 compilation (#5594)
  • Add missing validation for input count in argument promotion (#5592)
  • Don't return pointers to a local variable in dali_operator_test. (#5585)
  • Fix operator trace caching (#5580)
  • Fix readlink usage - readlink doens't null-terminate strings. (#5577)
  • Fix WarpPerspective::GetFillValue (#5575)
  • Prevent stack-use-after-scope (#5572)
  • Add missing #include <optional> in nvcvop.h (#5570)
  • Fix MakeContiguous sample_dim for empty batches. (#5567)
  • Set affinity by device UUID. (#5566)
  • Unchecked return value from CUDA library (#5564)

Breaking API changes

  • DALI 1.39 was the final release to support the MXNet integration.

Deprecated features

No features were deprecated in this release.

Known issues:

  • The following operators: experimental.readers.fits, experimental.decoders.video, and experimental.inputs.video 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.
  • 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 plugin might not be compatible with TensorFlow versions 1.15.0 and later.
    To use DALI with the TensorFlow version that does not have a prebuilt plugin binary shipped with DALI, make sure that the compiler that is used to build TensorFlow exists on the system during the plugin 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 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.

Binary builds

NOTE: DALI builds for CUDA 12 dynamically link the CUDA toolkit. To use DALI, install the latest CUDA toolkit.

CUDA 11.0 and CUDA 12.0 builds use CUDA toolkit enhanced compatibility. 
They are built with the latest CUDA 11.x/12.x toolkit respectively but they can run on the latest, 
stable CUDA 11.0/CUDA 12.0 capable drivers (450.80 or later and 525.60 or later respectively).
However, using the most recent driver may enable additional functionality. 
More details can be found in enhanced CUDA compatibility guide.

Install via pip for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda120==1.41.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda120==1.41.0

or just:

pip install nvidia-dali-cuda120==1.41.0
pip install nvidia-dali-tf-plugin-cuda120==1.41.0

For CUDA 11:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-cuda110==1.41.0
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/ nvidia-dali-tf-plugin-cuda110==1.41.0

or just:

pip install nvidia-dali-cuda110==1.41.0
pip install nvidia-dali-tf-plugin-cuda110==1.41.0

Or use direct download links (CUDA 12.0):

Or use direct download links (CUDA 11.0):

FFmpeg source code:

  • This software uses code of FFmpeg licensed under the LGPLv2.1 and its source can be downloaded here

Libsndfile source code:

Loading