\brief Returns the size (in bytes) of the buffer.

Get the NumPy array, deserializing from the disk if it was stored earlier. Returns: np.ndarray: The NumPy array

Power implementation. \param base value to exponentiate \param exp power to expontiate to \return function value stored in single-preicision

Removes an entire folder if folder is empty. Provides print info statements. Args: fpath: Location to folder success_msg: Su

\brief Set allowed formats for this tensor. By default all formats are allowed. Shape tensors (for which isShapeTensor() returns true) may only have r

\brief Set the padding mode. Padding mode takes precedence if both setPaddingMode and setPre/PostPadding are used. Default: kEXPLICIT_ROUND_DOWN \s

Set plugin namespace

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

\brief Set the minimum / optimum / maximum values for an input shape tensor. This function must be called three times for every input tensor t that i

\brief Set the stride for computing the output slice data. \param stride The dimensions of the stride to compute the values to store in the output sl

Args: value: An instance of :class:`QuantDescriptor <pytorch_quantization.tensor_quant.QuantDescriptor>`

Used to cache encoder hidden state runs across same encoder sessions

Set the input metadata for the data loader. Args: input_metadata (TensorMetadata): Input Metadata, i

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

Creates a function that applies a Top-K operation to a IterationResult. Top-K will return the indices of the k largest values in the

White background with colored grid

Helper function to compute unpadded volume of a Dims (1 if 0 dimensional)

A TensorRT algorithm selector that can replay tactics according to a tactic replay file. Args: replay (Union[path, file-like, Tactic

Get a previously generated ONNX node (skip those that were not generated). Target index can be passed for jumping to a specific index.

Add a single entry for a single error in a subgraph. Multiple node ranges may apply to a single op/error combination. Args:

\brief Add a Scale layer to the network. \param input The input tensor to the layer. This tensor is required to have a minimum of 3 dimensions in imp

Converts current model into an ONNX model. Uses provided converter to convert object or default_convert is used instead if available.

Copy the graph. This makes copies of all nodes and tensors in the graph, but will not do a deep-copy of weights or attribute

\brief Query whether dynamic range is set. \return True if dynamic range is set, false otherwise.

Floor implementation. \param arg value to round \return rounded value

Creates a Polygraphy ``Algorithm`` instance from a TensorRT ``IAlgorithmContext`` and ``IAlgorithm``. Args: cont

Generates a name based on two string inputs, and checks if the combination is valid.

Creates a native tf.keras ResNet model. Args: resnet_depth (str): ResNet depth. Options=[50 (default), 101, 152]. resnet_ver

Returns whether this tensor is considered empty in the graph. *Note: 'Empty' here refers to the name of the tensor, which is omitted

Classification implementation. \param arg value to classify \retval true if infinite number \retval false else

Natural logarithm of gamma function. \param arg function argument \return natural logarith of gamma function for \a arg template<typename T> typename

Natural logarithm of gamma function. \param arg function argument \return natural logarith of gamma function for \a arg template<typename T> typename

Loads a TensorRT engine according to arguments provided on the command-line. Args: network (Tuple[trt.Builder, trt.INetw

Partitions a set of subgraphs into supported and unsupported subgraphs. Args: nodelists (List[Tuple[List[int

Registers a dictionary of all available argument groups with this argument group. Args: arg_groups (ArgGroups): Maps arg

A special fixture that runs commands, but sandboxes any `pip install`s in a virtual environment. Packages from the test environment are still

Sanitize the graph by cleaning any unconnected nodes, do a topological resort, and fold constant inputs values. When possible, run sh

Sanitize the graph by cleaning any unconnected nodes, do a topological resort, and fold constant inputs values. When possible, run sh

Ensures that every tensor used by a network has a dynamic range set. All tensors in a network must have a dynamic range specified if a calibrator is

\brief Set the LRN alpha value. The valid range is [-1e20, 1e20]. \see getAlpha()

\brief Set the dynamic dimensions of an input binding. \param bindingIndex index of an input tensor whose dimensions must be compatible with the netw

\brief Set the ErrorRecorder for this interface Assigns the ErrorRecorder to this interface. The ErrorRecorder will track all errors during execution

\brief Set the padding for pooling. Default: 0 If executing this layer on DLA, both height and width of padding must be in the range [0,7]. \see ge