Installation¶
Prerequisites¶
Linux x64.
NVIDIA Driver supporting CUDA 10.0 or later (i.e., 410.48 or later driver releases).
(Optional) One or more of the following deep learning frameworks:
MXNet 1.3
mxnet-cu100
or later.PyTorch 0.4 or later.
TensorFlow 1.7 or later.
DALI in NGC Containers¶
DALI is preinstalled in the TensorFlow, PyTorch, and MXNet containers in versions 18.07 and later on NVIDIA GPU Cloud.
pip - Official Releases¶
nvidia-dali¶
Execute the following command to install the latest DALI for specified CUDA version (please check support matrix to see if your platform is supported):
for CUDA 11.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist --upgrade nvidia-dali-cuda110
for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist --upgrade nvidia-dali-cuda120
Note
CUDA 11.0 and CUDA 12.0 build uses CUDA toolkit enhanced compatibility. It is built with the latest CUDA 11.x/12.x respectively toolkit while it can run on the latest, stable CUDA 11.0 and CUDA 12.0 capable drivers (450.80 or later and 525.60 or later respectively). Using the latest driver may enable additional functionality. More details can be found in enhanced CUDA compatibility guide.
Note
Please always use the latest version of pip available (at least >= 19.3) and update when possible by issuing pip install –upgrade pip
nvidia-dali-tf-plugin¶
DALI doesn’t contain prebuilt versions of the DALI TensorFlow plugin. It needs to be installed as a separate package which will be built against the currently installed version of TensorFlow:
for CUDA 11.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist --upgrade nvidia-dali-tf-plugin-cuda110
for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist --upgrade nvidia-dali-tf-plugin-cuda120
Installing this package will install nvidia-dali-cudaXXX
and its dependencies, if they are not already installed. The package tensorflow-gpu
must be installed before attempting to install nvidia-dali-tf-plugin-cudaXXX
.
Note
The packages nvidia-dali-tf-plugin-cudaXXX
and nvidia-dali-cudaXXX
should be in exactly the same version.
Therefore, installing the latest nvidia-dali-tf-plugin-cudaXXX
, will replace any older nvidia-dali-cudaXXX
version already installed.
To work with older versions of DALI, provide the version explicitly to the pip install
command.
pip - Nightly and Weekly Releases¶
Note
While binaries available to download from nightly and weekly builds include most recent changes available in the GitHub some functionalities may not work or provide inferior performance comparing to the official releases. Those builds are meant for the early adopters seeking for the most recent version available and being ready to boldly go where no man has gone before.
Note
It is recommended to uninstall regular DALI and TensorFlow plugin before installing nightly or weekly builds as they are installed in the same path
Nightly Builds¶
To access most recent nightly builds please use flowing release channel:
for CUDA 11.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/nightly --upgrade nvidia-dali-nightly-cuda110
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/nightly --upgrade nvidia-dali-tf-plugin-nightly-cuda110
for CUDA 12.0:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/nightly --upgrade nvidia-dali-nightly-cuda120
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/nightly --upgrade nvidia-dali-tf-plugin-nightly-cuda120
Weekly Builds¶
Also, there is a weekly release channel with more thorough testing. To access most recent weekly builds please use the following release channel (available only for CUDA 12):
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/weekly --upgrade nvidia-dali-weekly-cuda120
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/weekly --upgrade nvidia-dali-tf-plugin-weekly-cuda120
pip - Legacy Releases¶
For older versions of DALI (0.22 and lower), use the package nvidia-dali. The CUDA version can be selected by changing the pip index:
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/cuda/9.0 --upgrade nvidia-dali
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/cuda/9.0 --upgrade nvidia-dali-tf-plugin
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/cuda/10.0 --upgrade nvidia-dali
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/cuda/10.0 --upgrade nvidia-dali-tf-plugin
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist --upgrade nvidia-dali-cuda102
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist --upgrade nvidia-dali-tf-plugin-cuda102
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/cuda/11.0 --upgrade nvidia-dali
pip install --extra-index-url https://developer.download.nvidia.com/compute/redist/cuda/11.0 --upgrade nvidia-dali-tf-plugin
CUDA 11 build is provided starting from DALI 0.22.0.
CUDA 10.2 build is provided starting from DALI 1.4.0 up to DALI 1.20.
CUDA 10 build is provided up to DALI 1.3.0.
CUDA 9 build is provided up to DALI 0.22.0.
Open Cognitive Environment (Open-CE)¶
DALI is also available as a part of the Open Cognitive Environment - a project that contains everything that is needed to build conda packages for a collection of machine learning and deep learning frameworks.
This effort is community-driven and the DALI version available there may not be up to date.
Prebuild packages (including DALI) are hosted by external organizations.