* @brief Number of samples */

Wrapper combining unittest assertRaises and assertRaisesRegex. Specify ``regex=pattern`` or ``glob=pattern`` to check error message of expect

* @brief Returns the extent of the region of interest as (end - begin) */

* @brief Returns the number of inputs. */

Set the outputs of the pipeline. Use of this function is an alternative to overriding `define_graph` in a derived class. Args

Returns the run-time shapes of this DataNode as a new DataNode Parameters ---------- arg_dtype : DALIDataType, optional

* @brief Returns the id of the datatype of the underlying storage. */

* @brief Returns the CUDA stream that this work is to be done in. */

Run the pipeline and return the result on the specified CUDA stream. If the pipeline was created with `exec_pipelined` option set to

Build the pipeline (optional step). Instantiates the pipeline's backend objects and starts processing threads. If the pipeline uses m

Computes maxima of corresponding values in `left` and `right`. :rtype: TensorList of the type that is calculated based on the type promotion rule

* @brief Trims the current region to `tail_start` blocks and returns the rest as a new region. * * @note This function operates purely on me

Pass a multidimensional array or DLPack (or a list thereof) to an eligible operator. The operators that may be provided with data using this

Add a reference from operator to this notebook with specified docstring. Parameters ---------- operator : str Name of operator wi

* @brief Get the runtime information about the type of the sample */

* @brief true if there is a value explicitly provided (constant or argument input) */

Resolves ambiguity when used with brace-enclosed initializer list

Returns the concrete results of the invocation. The invocation may have multiple results (e.g. readers may produce data + labels).

Compare the outputs of two pipelines across several iterations. Args: pipe1: input pipeline object. pipe2: input pipeline object.

* @brief Return a const, un-typed pointer to the underlying storage. */

Compare two batches of data, be it dali TensorList or list of numpy arrays. Args: batch1: input batch batch2: input batch

Returns a tuple that enables all options.

Resize already allocated shared memory chunk. If you want to resize the underlying shared memory chunk pass trunc=True, if the memory chunk ha

Computes minima of corresponding values in `left` and `right`. :rtype: TensorList of the type that is calculated based on the type promotion rule