jet.bindings.TensorC64¶

class TensorC64¶

Bases: pybind11_builtins.pybind11_object

Tensor represents an \(n\)-rank data structure of complex-valued data for tensor operations. We use the following conventions:

“Rank” and “order” are used interchangeably and refer to the number of tensor indices. In general, “rank” will be preferred.
“Dimension” refers to the number of elements along a tensor index.
“Shape” refers to the dimensions of a tensor; the number of dimensions is the rank of the tensor.

Parameters

indices (Sequence[str]) – Label of each tensor index.
shape (Sequence[int]) – Dimension of each tensor index.
data (Sequence[complex]) – Row-major encoded complex data representation.

Attributes

`data`	Complex data values in row-major order.
`dtype`
`index_to_dimension_map`	Mapping from index labels to dimension sizes.
`indices`	List of index labels.
`scalar`	First data value of the tensor.
`shape`	List containing the dimension of each tensor index.

data¶: Complex data values in row-major order.

dtype¶

index_to_dimension_map¶: Mapping from index labels to dimension sizes.

indices¶: List of index labels.

scalar¶: First data value of the tensor.

shape¶: List containing the dimension of each tensor index.

Methods

`add_tensor`(self, other)	Adds two tensor objects with the same index sets.
`conj`(self)	Returns the conjugate of the given tensor object.
`contract_with_tensor`(self, other)	Contracts two tensor objects over the intersection of their index sets.
`fill_random`(args, *kwargs)	Overloaded function.
`get_value`(self, indices)	Returns the tensor data value at the given \(n\)-dimensional index.
`init_indices_and_shape`(self, indices, shape)	Initializes the indices and shape of a tensor object.
`is_scalar`(self)	Reports whether the tensor is a scalar.
`rename_index`(self, pos, new_label)	Renames the index label at the given position.
`reshape`(self, shape)	Reshapes a tensor object to the given dimensions.
`set_value`(self, indices, value)	Sets the tensor data value at the given \(n\)-dimensional index.
`slice_index`(self, index, value)	Slices a tensor object index.
`transpose`(args, *kwargs)	Overloaded function.

add_tensor(self: jet.bindings.TensorC64, other: jet.bindings.TensorC64) → jet.bindings.TensorC64 ¶

Adds two tensor objects with the same index sets. The resulting tensor will have the same indices as the first argument (i.e., A).

Parameters

A (Tensor) – Tensor on the LHS of the addition.
B (Tensor) – Tensor on the RHS of the addition.

Returns

Element-wise sum of the tensors.

Return type

Tensor

Example

Given a 2x3 tensor \(A(i,j)\) and a 2x3 tensor \(B(i,j)\), the addition of \(A\) and \(B\) is another 2x3 tensor \(C(i,j)\):

import jet

A = jet.Tensor(["i", "j"], [2, 3])
B = jet.Tensor(["i", "j"], [2, 3])

A.fill_random()
B.fill_random()

C = A.add_tensor(B);

conj(self: jet.bindings.TensorC64) → jet.bindings.TensorC64 ¶

Returns the conjugate of the given tensor object.

Returns: Conjugate tensor object.
Return type: Tensor

contract_with_tensor(self: jet.bindings.TensorC64, other: jet.bindings.TensorC64) → jet.bindings.TensorC64 ¶

Contracts two tensor objects over the intersection of their index sets. The resulting tensor will be formed with indices given by the symmetric difference of the index sets.

Parameters

A (Tensor) – Tensor on the LHS of the contraction.
B (Tensor) – Tensor on the RHS of the contraction.

Returns

Contraction of the tensors.

Return type

Tensor

Example

Given a 3x2x4 tensor \(A(i,j,k)\) and a 2x4x2 tensor \(B(j,k,l)\), the common indices are \(\{j,k\}\) and the symmetric difference of the sets is \(\{i,l\}\). The result of the contraction will be a tensor 3x2 tensor \(C(i,l)\).

import jet

A = jet.Tensor(["i", "j", "k"], [3, 2, 4])
B = jet.Tensor(["j", "k", "l"], [2, 4, 2])

A.fill_random()
B.fill_random()

C = A.contract_with_tensor(B);

fill_random(*args, **kwargs)¶

Overloaded function.

fill_random(self: jet.bindings.TensorC64) -> None

Assigns random values to the tensor data. The real and imaginary components of each datum will be independently sampled from a uniform distribution with support over \([-1, 1]\).

fill_random(self: jet.bindings.TensorC64, seed: int) -> None

Assigns random values to the tensor data. The real and imaginary components of each datum will be independently sampled from a uniform distribution with support over \([-1, 1]\). This overload enables reproducible random number generation for a given seed.

Parameters: seed (int) – Seed to supply to the RNG engine.

get_value(self: jet.bindings.TensorC64, indices: List[int]) → complex¶

Returns the tensor data value at the given \(n\)-dimensional index.

Parameters: indices (Sequence[str]) – \(n\)-dimensional tensor data index in row-major order.
Returns: Complex data value at the given index.
Return type: complex

init_indices_and_shape(self: jet.bindings.TensorC64, indices: List[str], shape: List[int]) → None¶

Initializes the indices and shape of a tensor object. The indices and shape must be ordered to map directly such that indices[i] has size shape[i].

Parameters

indices (Sequence[str]) – Label of each tensor index.
shape (Sequence[int]) – Dimension of each tensor index.

is_scalar(self: jet.bindings.TensorC64) → bool¶

Reports whether the tensor is a scalar.

Returns: True if the tensor is of rank 0. Otherwise, False is returned.
Return type: bool

rename_index(self: jet.bindings.TensorC64, pos: int, new_label: str) → None¶

Renames the index label at the given position.

Parameters

pos (int) – Position of the label.
new_label (str) – New label.

reshape(self: jet.bindings.TensorC64, shape: List[int]) → jet.bindings.TensorC64 ¶

Reshapes a tensor object to the given dimensions.

Parameters: shape (Sequence[int]) – Index dimensionality of the reshaped tensor object.
Returns: Reshaped copy of the given tensor object.
Return type: Tensor

set_value(self: jet.bindings.TensorC64, indices: List[int], value: complex) → None¶

Sets the tensor data value at the given \(n\)-dimensional index.

Parameters

indices (Sequence[str]) – \(n\)-dimensional tensor data index in row-major order.
value (complex) – Value to set at the data index.

slice_index(self: jet.bindings.TensorC64, index: str, value: int) → jet.bindings.TensorC64 ¶

Slices a tensor object index. The result is a tensor object whose indices and data are a subset of the provided tensor object, sliced along the given index argument.

Parameters

index (str) – Index label on which to slice.
value (int) – Value to slice the index on.

Returns

Slice of the tensor object.

Return type

Tensor

Example

Suppose that \(A(i,j)\) is a 2x3 tensor. Then,

import jet

A = jet.Tensor({"i", "j"}, {2, 3})
A.fill_random()

A.slice_index("i", 0) # Result is a 1x3 tensor
A.slice_index("i", 1) # Result is a 1x3 tensor

A.slice_index("j", 0) # Result is a 2x1 tensor
A.slice_index("j", 1) # Result is a 2x1 tensor
A.slice_index("j", 2) # Result is a 2x1 tensor

transpose(*args, **kwargs)¶

Overloaded function.

transpose(self: jet.bindings.TensorC64, new_indices: List[str]) -> jet.bindings.TensorC64

Transposes the indices of a tensor object.

Parameters: indices (Sequence[str]) – Desired index ordering, specified as a list of labels.
Returns: Transposed tensor object.
Return type: Tensor

transpose(self: jet.bindings.TensorC64, new_ordering: List[int]) -> jet.bindings.TensorC64

Transposes the indices of a tensor object.

Parameters: ordering (Sequence[int]) – Desired index ordering, specified as a permutation.
Returns: Transposed tensor object.
Return type: Tensor

code/api/jet.bindings.TensorC64

Download Python script

Download Notebook

View on GitHub

jet.bindings.TensorC64¶

Attributes

Methods

Contents