Converts a given path to a path relative to the Polygraphy root module. If the path is not part of the Polygraphy module, returns a p

Run argument parsing for all argument groups. Args: args (Namespace): The namespace returned by ``pa

\brief Report that a test has started. \pre reportTestStart() has not been called yet for the given testAtom \param[in] testAtom The handle to the t

Save the ONNX model to the given location. :param output_path: Path pointing to the location where to write out the updated ONNX mode

\brief Set the axis parameter. \see IOneHotLayer

\brief Sets custom dynamic range for network tensors

\brief Set the operation for the layer. \see getOperation(), TopKOperation

\brief Set the data type of a tensor. \param type The data type of the tensor. The type is unchanged if the tensor is not a network input tensor, or

Square root. \param arg function argument \return square root of \a arg template<typename T> typename enable<expr,T>::type sqrt(T arg) { return functi

Logs a message to stdout with ULTRA_VERBOSE severity. Args: message (Union[str, Callable() -> str]):

Saves an engine to the provided path. Args: engine (Union[trt.ICudaEngine, Callable() -> trt.ICudaEngine]):

Create the experiment directory (sub-folder in the base directory passed to this class). Args: folder_name (str): name o

Adds a runner to the script. Args: runner_str (str): A string constructing the runner.

\brief Determine whether a binding is an input binding. \param bindingIndex The binding index. \return True if the index corresponds to an input bind

\brief Returns pointer to underlying array.

Ensures that the input is marked as a safe string (i.e. Script.String(safe=True)).

Frees all device buffers associated with this calibrator

Dummy network with resnet like architecture.

\brief Determine the required data type for a buffer from its binding index. \param bindingIndex The binding index. \return The type of the data in t

Return true if expression is a build-time constant.

Check for NaN. \param arg number to check \retval true for NaNs \retval false else template<typename T> typename enable<bool,T>::type isnan(T arg) { r

Check for NaN. \param arg number to check \retval true for NaNs \retval false else template<typename T> typename enable<bool,T>::type isnan(T arg) { r

Classification implementation. \param arg value to classify \retval true if not a number \retval false else

Helper function to create inline TRT enums for usage across various TRT classes.

Returns a context manager that supplies unique integer IDs for Nodes in the Graph. For example: :: with graph.n

\brief Populates the Args struct with the provided command-line parameters. \throw invalid_argument if any of the arguments are not valid \return bo

Nearest integer implementation. \param arg value to round \return rounded value

\brief Method for controlling the verbosity of logging output \param severity The logger will only emit messages that have severity of this level or

Hyperbolic tangent implementation. \param arg function argument \return function value stored in single-preicision

Modifies this tensor in-place to convert it to a Constant. This means that all consumers/producers of the tensor will see the update.

Creates the initializers with weights from the weights file together with the input tensors. Keyword arguments: conv_params -

Get input Tensors parent recursively. Here we want to know id DequantizeLinear is tensors parent. Recursively we go up the gr

Apply quantization on input and weight Function called by the classes lower in the hierarchy, which actually performs the quantization before

Add arguments to the command-line parser. This will also add arguments for any subtools reported by `get_subtools()`. Args:

\brief Create a new optimization profile. If the network has any dynamic input tensors, the appropriate calls to setDimensions() must be made. Likewi

\brief Synchronously execute inference a network. This method requires an array of input and output buffers. The mapping from tensor names to indices

Exponential function. \param arg function argument \return e raised to \a arg template<typename T> typename enable<expr,T>::type exp(T arg) { return f

Resets parameter data pointer so that they can use faster code paths. Right now, this works only if the module is on the GPU and cuDNN is ena

Nearest integer not greater than half value. \param arg half to round \return nearest integer not greater than \a arg template<typename T> typename en

\brief Get the value of the axis parameter.

\brief Get the minimum / optimum / maximum dimensions for a particular input binding under an optimization profile. \param bindingIndex The input bin

\brief Return strides of the input/output tensor of algorithm. For vectorized formats, strides are given in units of vectors. \return the strides of

\brief Get the weights for the layer. \see setWeights

Return dictionary with information about layers to quantize for quantize config object. Args: None Return

Get TimingProfile settings given current args.

Run shape inference on an ONNX model if `do_shape_inference` is True according to arguments provided on the command-line. Ar

\brief Whether the tensor is a shape tensor. A shape tensor is a tensor that is related to shape calculations. It must have type Int32, Bool, or Floa

Multiply by power of two. \param arg number to modify \param exp power of two to multiply with \return \a arg multplied by 2 raised to \a exp template

Multiply by power of two. \param arg number to modify \param exp power of two to multiply with \return \a arg multplied by 2 raised to \a exp template

Similar to `make_invocable`, but will return ``None`` if all arguments are ``None``. For example: :: >>> make_invocable_if_nond