Configuration for a single operator test.

Composes transforms together in a single pipeline This class chains multiple DALI operations in a sequential manner, similar to ``torchv

Describes a callback to be used as a per-sample/per-frame argument to the operator. ---------- name : Union[str, int] String with

Represents a qualified name.

Temporary half-precision expression. This class represents a half-precision expression which just stores a single-precision value internally.

Randomly horizontally flips the given image randomly with a given probability. Parameters ---------- p : float Probability o

DALI Dynamic Mode data type. This class is used to represent the data type of a Tensor or Batch. Data types such as ``uint32`` are insta

A PyTorch DataLoader wrapper with in-memory caching and performance monitoring. This class wraps a standard PyTorch DataLoader and adds:

Indicates that this module should be not converted.

Class that holds a version string accompanied with maximum and minimum python version that this version should support. If python falls beyond ver

Device on which data is stored and operators are executed. The device can be either CPU or (specific) GPU. .. warning:: It's re

A Tensor object. This class represents a single tensor usable with DALI dynamic API. It can contain any of the following: - tensor data

NVTX range marker for the DALI domain. Can be used as a context manager or decorator. Use categories to organize annotations.

.. note:: This class should not be instantiated directly; use :func:`Constant` function with appropriate arguments to create inst

A Batch object. This class represents a batch of tensors usable with DALI dynamic API. The tensors in the batch have the same element type, l

Convert images or videos to grayscale. Parameters ---------- num_output_channels : int Number of channels desired for output

Signals ``PipelineHWC`` to return a uint8 CHW ``torch.Tensor`` instead of ``PIL.Image``. When placed in a ``Compose`` op list this operator

Resize the input image to the given size If the image is torch Tensor, it is expected to have […, H, W] shape, where … means a maximum of

Describes the indices of a sample requested from :meth:`nvidia.dali.fn.external_source` :ivar idx_in_epoch: 0-based index of the sample with

Describes a batch or sub-batch of work in sample mode that consists of SampleInfo instances with consecutive indices. It denotes range of sam

An exception thrown when an invalid dictionary key is provided * * The exception denotes an invalid key. It can be thrown when: * - the key is not

Computes and stores the average and current value

Reference to one output of a compile graph node.

DALI data loader to be used in the main loop, which replaces the pipeline run references with actual data produced by the DALI server. Se

Generates inputs of required shapes and types The number of inputs is based on the length of tuple `types`, if types is a single element

Crop the input at a random location. If the input is a ``torch.Tensor`` it can have an arbitrary number of leading batch dimensions. For

Signals ``PipelineCHW`` to return a ``PIL.Image`` instead of ``torch.Tensor``. When placed in a ``Compose`` op list this operator does not m

Apply randomly a list of transformations with a given probability. Parameters ---------- op_list : Sequence[Callable] List o

Crop a random portion of the input and resize it to a given size. If the input is a ``torch.Tensor`` it can have an arbitrary number of lead

Signals the pipeline to return a ``torch.Tensor`` instead of the default output type. When placed in a ``Compose`` op list this operator doe

Normalize a tensor image with mean and standard deviation. This transform does not support PIL Image. Given mean: (mean[1],...,mean[n])

A container to keep running sum, mean and last value.

DALI iterator for classification tasks for PyTorch. It returns 2 outputs (data and label) in the form of PyTorch's Tensor. Calling

This class is a symbolic representation of a TensorList and is used at graph definition stage. It does not carry actual data, but is used to defin

Randomly vertically flips the given image randomly with a given probability. Parameters ---------- p : float Probability of

Simple fixed capacity shared memory queue of fixed size messages. Writing to a full queue fails, attempt to get from an empty queue blocks un

Computes and stores the average and current value

Wrapper class for a CUDA stream.

Blurs image with randomly chosen Gaussian blur kernel. The convolution will be using reflection padding corresponding to the ``kernel size``