Encode an object to JSON. NOTE: For Polygraphy objects, you should use the ``to_json()`` method instead. Returns: str: A JSON r

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

Creates a generator that yields runner outputs (List[IterationResult]).

Saves an ONNX model to the specified path. Args: model (Union[onnx.ModelProto, Callable() -> onnx.ModelProto]):

Since torch functions often allow for models to either be from disk as fpath or from a loaded object, we provide a similar option her

Returns the layer's class, module, and name

Use currying to return True/False specialized fns to the cond.

Absolute value. \param arg operand \return absolute value of \a arg template<typename T> typename enable<T,T>::type abs(T arg) { return unary_speciali

For the modules registered in the quant_map, simply monkey patch them and also store the original modules so that they could be later

Checks if 'current_layer_name' is in 'qspec'. It returns True if 'current_layer_name' is NOT in 'qspec' and False if it is. This means that

Fixture that ensures warnings are emitted when `_impl` methods of runners are called.

Clean quantize config object from global space. Calling this is important to use `quantize_model` multiple times within a single modu

Converts a sequence of [tokens|ids] using the vocab.

Converts a trt shape spec to a trtexec shape spec

\brief Create timing cache Create ITimingCache instance from serialized raw data. The created timing cache doesn’t belong to a specific IBuilderConfi

Serialize to dict The build-in __dict__ method returns all the attributes, which includes those have default value and have protected

Create directory is doesn't exist already. Args: dirname: Name of the directory to create.

Fill this profile with sane default values for any bindings whose shapes have not been set explicitly. Args: net

Helper function to fine-tune QAT model. Args: q_model (tf.keras.Model): Keras model. train_batches (tf.data.Dataset): train

Remainder of division. \param x first operand \param y second operand \return remainder of floating point division. template<typename T,typename U> ty

Remainder of division. \param x first operand \param y second operand \return remainder of floating point division. template<typename T,typename U> ty

Freeze this context so that the `state` and `iteration_info` cannot be modified.

Constructs a new TensorMetadata using information from the provided feed_dict. Args: feed_dict (OrderedDict[str, numpy.n

Decode a JSON string to an object. NOTE: For Polygraphy objects, you should use the ``from_json()`` method instead. Args: src (

\brief Get description of all weights that could be refit. \param size The number of items that can be safely written to a non-null layerNames or rol

\brief Return the number of components included in one element. The number of elements in the vectors is returned if getBindingVectorizedDim() != -1.

\brief Return the dimension index that the buffer is vectorized, or -1 is the name is not found. Specifically -1 is returned if scalars per vector is

\brief get the ErrorRecorder assigned to this interface. Retrieves the assigned error recorder object for the given class. A nullptr will be returned

Util to get the full path to the downloaded data files. It only works when the sample doesn't have any other command line argument.

\brief Get description of missing weights. For example, if some Weights have been set, but the engine was optimized in a way that combines weights, a

\brief get the computational precision of this layer \return the computational precision \see setPrecision() precisionIsSet() resetPrecision()

\brief Return the number of components included in one element, or -1 if the provided name does not map to an input or output tensor. The number of e

\brief Determine whether a tensor is an input or output tensor. \param tensorName The name of an input or output tensor. \return kINPUT if tensorNam

\brief Get shape of an input or output tensor. \param tensorName The name of an input or output tensor. \return shape of the tensor, with -1 in plac

\brief Return the dimension index that the buffer is vectorized, or -1 if the provided name does not map to an input or output tensor. Specifically -

Gets the global logger after applying command-line options. Returns: G_LOGGER

Create unique plans names

Get the specs for the input tensor of the network. Useful to prepare memory allocations. :return: Two items, the shape of the input t

Get the specs for the input tensor of the network. Useful to prepare memory allocations. :return: Two items, the shape of the input t

Whether we are currently on a line between the specified start and end marker. This will always return False for a line containing th

Returns the total inference time in seconds required during the last call to ``infer()``. Must be called only after ``activate()`` a

Search for dangling Q/DQ layers.

Helper for loading a specific script for given network.

Copies data between host and device memory. Args: dst (int): The memory address of the destination,

Whether this model needs to be loaded via a Polygraphy ONNX loader, e.g., in case it needs modifications. Args:

Lower text and remove punctuation, articles and extra whitespace.

Lower text and remove punctuation, articles and extra whitespace.

Converting variables in the onnx checkpoint to names corresponding to the naming convention used in the TF version, expected by the builder