Today’s deep learning applications include complex, multi-stage pre-processing data pipelines that include compute-intensive steps mainly carried out on the CPU. For instance, steps such as load data from disk, decode, crop, random resize, color and spatial augmentations and format conversions are carried out on the CPUs, limiting the performance and scalability of training and inference tasks. In addition, the deep learning frameworks today have multiple data pre-processing implementations, resulting in challenges such as portability of training and inference workflows and code maintainability.
NVIDIA Data Loading Library (DALI) is a collection of highly optimized building blocks and an execution engine to accelerate input data pre-processing for deep learning applications. DALI provides both performance and flexibility of accelerating different data pipelines, as a single library, that can be easily integrated into different deep learning training and inference applications.
Key highlights of DALI include:
- Full data pipeline accelerated from reading from disk to getting ready for training/inference
- Flexibility through configurable graphs and custom operators
- Support for image classification and segmentation workloads
- Ease of integration through direct framework plugins and open source bindings
- Portable training workflows with multiple input formats - JPEG, LMDB, RecordIO, TFRecord
- Extensible for user specific needs through open source license
DALI is preinstalled in the NVIDIA GPU Cloud TensorFlow, PyTorch, and MXNet containers in versions 18.07 and later.
- Linux x64
- NVIDIA Driver_ supporting CUDA 9.0 or later (i.e., 384.xx or later driver releases)
- One or more of the following deep learning frameworks:
- MXNet 1.3_
mxnet-cu90
or later - PyTorch 0.4_
- TensorFlow 1.7_ or later
- MXNet 1.3_
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist nvidia-dali
Note
nvidia-dali package contains prebuilt versions of the dali tensorflow plugin for several versions of tensorflow. Since release 0.6.1 there is also a possibility to install dali tensorflow plugin for the currently installed version of tensorflow, thus allowing forward compatibility:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist nvidia-dali-tf-plugin
Note
Installing this package will install nvidia-dali and its dependencies if not already installed. The package tensorflow-gpu must be installed before attempting to install nvidia-dali-tf-plugin.
Note
The package nvidia-dali-tf-plugin has a strict requirement with nvidia-dali as its exact same version. Thus, installing nvidia-dali-tf-plugin at its latest version will replace any older nvidia-dali versions already installed with the latest. To work with older versions of DALI, please provide the version explicitely to the pip install command.
OLDER_VERSION=0.6.1
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist nvidia-dali-tf-plugin==$OLDER_VERSION
Linux x64 | |
GCC 4.9.2_ or later | |
Boost 1.66_ or later | Modules: preprocessor |
GCC 4.9.2_ or later NVIDIA CUDA 9.0_ | CUDA 8.0 compatibility is provided unofficially |
nvJPEG library_ | This can be unofficially disabled. See below |
protobuf_ | version 2 or later (version 3 or later is required for TensorFlow TFRecord file format support) |
CMake 3.5_ or later | |
libjpeg-turbo 1.5.x_ or later | This can be unofficially disabled. See below |
FFmpeg 3.4.2_ or later | We recommend using version 3.4.2 compiled following the instructions below. |
OpenCV 3_ or later | We recommend using version 3.4+, however previous versions are also compatible. OpenCV 2.x compatibility is provided unofficially |
(Optional) liblmdb 0.9.x_ or later | |
|
ng frameworks: later |
Note
TensorFlow installation is required to build the TensorFlow plugin for DALI
Note
Items marked "unofficial" are community contributions that are believed to work but not officially tested or maintained by NVIDIA.
Note
This software uses code of FFmpeg licensed under the LGPLv2.1 and its source can be downloaded https://developer.download.nvidia.com/compute/redist/nvidia-dali/ffmpeg-3.4.2.tar.bz2
FFmpeg was compiled using the following command line:
./configure \
--prefix=/usr/local \
--disable-static \
--disable-all \
--disable-autodetect \
--disable-iconv \
--enable-shared \
--enable-avformat \
--enable-avcodec \
--enable-avfilter \
--enable-protocol=file \
--enable-demuxer=mov,matroska \
--enable-bsf=h264_mp4toannexb,hevc_mp4toannexb && \
make
git clone --recursive https://github.com/NVIDIA/dali
cd dali
mkdir build
cd build
To build DALI without LMDB support:
cmake ..
make -j"$(nproc)"
To build DALI with LMDB support:
cmake -DBUILD_LMDB=ON ..
make -j"$(nproc)"
To build DALI using Clang (experimental):
Note
This build is experimental and it is not maintained and tested like the default configuration. It is not guaranteed to work. We recommend using GCC for production builds.
cmake -DCMAKE_CXX_COMPILER=clang++ -DCMAKE_C_COMPILER=clang ..
make -j"$(nproc)"
Optional CMake build parameters:
BUILD_PYTHON
- build Python bindings (default: ON)BUILD_TEST
- include building test suite (default: ON)BUILD_BENCHMARK
- include building benchmarks (default: ON)BUILD_LMDB
- build with support for LMDB (default: OFF)BUILD_NVTX
- build with NVTX profiling enabled (default: OFF)BUILD_TENSORFLOW
- build TensorFlow plugin (default: OFF)WERROR
- treat all build warnings as errors (default: OFF)- (Unofficial)
BUILD_JPEG_TURBO
- build with libjpeg-turbo (default: ON) - (Unofficial)
BUILD_NVJPEG
- build with nvJPEG (default: ON)
pip install dali/python
Linux x64 | |
Docker_ | Please follow installation guide and manual there |
Enter Docker directory and run ./build.sh. If needed, set the following environment variables:
- PYVER - Python version, default is 2.7
- CUDA_VERSION - version of the CUDA toolkit, default is 10
- NVIDIA_BUILD_ID - custom ID of the build, default is 1234
- CREATE_RUNNER - create Docker image with cuDNN, CUDA and DALI installed inside. It will create the Docker_run_cuda image, which needs to be run using nvidia-docker and DALI wheel in the wheelhouse directory under DALI/, default is NO
- CREATE_WHL - create a wheel as well, default is YES
The docs/examples
_ directory contains a series of examples (in the form of Jupyter notebooks) highlighting different features of DALI. It also contains examples of how to use DALI to interface with deep learning frameworks.
Documentation for the latest stable release is available here. Nightly version of the documentation that stays in sync with the master branch is available here.
- GPU Technology Conference 2018 presentation about DALI, T. Gale, S. Layton and P. Tredak: slides, recording.
Contributions to DALI are more than welcome. To contribute to DALI and make pull requests, follow the guidelines outlined in the Contributing document.
We appreciate any feedback, questions or bug reporting regarding this project. When help with code is needed, follow the process outlined in the Stack Overflow (https://stackoverflow.com/help/mcve) document. Ensure posted examples are:
- minimal – use as little code as possible that still produces the same problem
- complete – provide all parts needed to reproduce the problem. Check if you can strip external dependency and still show the problem. The less time we spend on reproducing problems the more time we have to fix it
- verifiable – test the code you're about to provide to make sure it reproduces the problem. Remove all other problems that are not related to your request/question.
DALI was built with major contributions from Trevor Gale, Przemek Tredak, Simon Layton, Andrei Ivanov, Serge Panev