onnx_graphsurgeon.ir.tensor.Tensor Class Reference

Inheritance diagram for onnx_graphsurgeon.ir.tensor.Tensor:

Collaboration diagram for onnx_graphsurgeon.ir.tensor.Tensor:

Public Member Functions
def	__init__ (self)
def	__setattr__ (self, name, value)
def	is_empty (self)
def	to_constant (self, np.ndarray values)
def	to_variable (self, np.dtype dtype=None, Sequence[Union[int, str]] shape=[])
def	i (self, tensor_idx=0, producer_idx=0)
def	o (self, consumer_idx=0, tensor_idx=0)
def	__str__ (self)
def	__repr__ (self)
def	__eq__ (self, other)

Public Attributes
	name
	values
	dtype
	shape

Static Public Attributes
int	DYNAMIC = -1

Private Attributes
	__class__

Detailed Description

Abstract base class for tensors in a graph

Constructor & Destructor Documentation

__init__()

def onnx_graphsurgeon.ir.tensor.Tensor.__init__

(

self

)

**This class is abstract and cannot be constructed directly.**

Member Function Documentation

__setattr__()

def onnx_graphsurgeon.ir.tensor.Tensor.__setattr__	(		self,
			name,
			value
	)

is_empty()

def onnx_graphsurgeon.ir.tensor.Tensor.is_empty

(

self

)

Returns whether this tensor is considered empty in the graph.

*Note: 'Empty' here refers to the name of the tensor, which is omitted for
optional tensors, NOT the shape of the tensor*

Returns:
    bool: Whether the tensor is empty, meaning that it is used for an omitted optional input or output.

to_constant()

def onnx_graphsurgeon.ir.tensor.Tensor.to_constant	(		self,
		np.ndarray	values
	)

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

Args:
    values (np.ndarray): The values in this tensor

Returns:
    self

Reimplemented in onnx_graphsurgeon.ir.tensor.Variable.

to_variable()

def onnx_graphsurgeon.ir.tensor.Tensor.to_variable	(		self,
		np.dtype	dtype = None,
		Sequence[Union[int, str]]	shape = []
	)

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

Args:
    dtype (np.dtype): The data type of the tensor.
    shape (Sequence[int]): The shape of the tensor.

Returns:
    self

Reimplemented in onnx_graphsurgeon.ir.tensor.Constant.

i()

def onnx_graphsurgeon.ir.tensor.Tensor.i	(		self,
			tensor_idx = 0,
			producer_idx = 0
	)

Convenience function to get an input tensor of one of this tensor's input nodes.
Note that the parameters are swapped compared to the o() function; this is because tensors are likely to have only a single producer

For example:
::

    assert tensor.i() == tensor.inputs[0].inputs[0]
    assert tensor.i(1, 2) == tensor.inputs[2].inputs[1]

Args:
    tensor_idx (int): The index of the input tensor of the input node. Defaults to 0.
    producer_idx (int): The index of the producer node of the input tensor, if the tensor has multiple producers. Defaults to 0.

Returns:
    Tensor: The specified producer (input) tensor.

o()

def onnx_graphsurgeon.ir.tensor.Tensor.o	(		self,
			consumer_idx = 0,
			tensor_idx = 0
	)

Convenience function to get an output tensor of one of this tensor's output nodes.

For example:
::

    assert tensor.o() == tensor.outputs[0].outputs[0]
    assert tensor.o(2, 1) == tensor.outputs[2].outputs[1]

Args:
    consumer_idx (int): The index of the consumer of the input tensor. Defaults to 0.
    tensor_idx (int): The index of the output tensor of the node, if the node has multiple outputs. Defaults to 0.

Returns:
    Tensor: The specified consumer (output) tensor

__str__()

def onnx_graphsurgeon.ir.tensor.Tensor.__str__

(

self

)

Here is the caller graph for this function:

__repr__()

def onnx_graphsurgeon.ir.tensor.Tensor.__repr__

(

self

)

Reimplemented in onnx_graphsurgeon.ir.tensor.Constant.

Here is the call graph for this function:

__eq__()

def onnx_graphsurgeon.ir.tensor.Tensor.__eq__	(		self,
			other
	)

Perform a check to see if two tensors are equal.

Tensors are considered equal if they share the same name. A Graph must not include Tensors with duplicate names.

Member Data Documentation

DYNAMIC

int onnx_graphsurgeon.ir.tensor.Tensor.DYNAMIC = -1

static

name

onnx_graphsurgeon.ir.tensor.Tensor.name

__class__

onnx_graphsurgeon.ir.tensor.Tensor.__class__

private

values

onnx_graphsurgeon.ir.tensor.Tensor.values

dtype

onnx_graphsurgeon.ir.tensor.Tensor.dtype

shape

onnx_graphsurgeon.ir.tensor.Tensor.shape

---

The documentation for this class was generated from the following file:

• tensor.py