Utils API Reference¶

This page provides the complete API reference for the batchtensor.utils module.

The utils module contains utility functions that support the main tensor operations, particularly for managing random seeds and ensuring reproducibility.

For usage examples and tutorials, see the Utils User Guide.

Seed Management¶

batchtensor.utils.seed ¶

Implements utility functions to manage random seeds for reproducible tensor operations.

batchtensor.utils.seed.get_random_seed ¶

get_random_seed(seed: int) -> int

Get a random seed.

Parameters:

Name	Type	Description	Default
`seed`	`int`	A random seed to make the process reproducible.	required

Returns:

Type	Description
`int`	A random seed. The value is between `-2 63` and `2 63 - 1`.

Example

>>> from batchtensor.utils.seed import get_random_seed
>>> get_random_seed(44)
6176747449835261347

batchtensor.utils.seed.get_torch_generator ¶

get_torch_generator(
    random_seed: int = 1,
    device: device | str | None = "cpu",
) -> Generator

Create a torch.Generator initialized with a given seed.

Parameters:

Name	Type	Description	Default
`random_seed`	`int`	A random seed.	`1`
`device`	`device \| str \| None`	The desired device for the generator.	`'cpu'`

Returns:

Type	Description
`Generator`	A `torch.Generator` object.

Example

>>> import torch
>>> from batchtensor.utils.seed import get_torch_generator
>>> generator = get_torch_generator(42)
>>> torch.rand(2, 4, generator=generator)
tensor([[0.8823, 0.9150, 0.3829, 0.9593],
        [0.3904, 0.6009, 0.2566, 0.7936]])
>>> generator = get_torch_generator(42)
>>> torch.rand(2, 4, generator=generator)
tensor([[0.8823, 0.9150, 0.3829, 0.9593],
        [0.3904, 0.6009, 0.2566, 0.7936]])

batchtensor.utils.seed.setup_torch_generator ¶

setup_torch_generator(
    generator_or_seed: int | Generator,
) -> Generator

Set up a torch.Generator object.

Parameters:

Name	Type	Description	Default
`generator_or_seed`	`int \| Generator`	A `torch.Generator` object or a random seed.	required

Returns:

Type	Description
`Generator`	A `torch.Generator` object.

Example

>>> from batchtensor.utils.seed import setup_torch_generator
>>> generator = setup_torch_generator(42)
>>> generator
<torch._C.Generator object at 0x...>