Constant

class seli.net.Constant(value)

Bases: Module

Return a constant value. Ignores all inputs.

Parameters:

value (jax.Array) – Constant value to return.

Attributes Summary

dtype	Return the dtype of the constant value.
shape	Return the shape of the constant value.

Methods Summary

__call__(*args, **kwargs)	Return the constant value, ignoring all inputs.
full([x, shape])	Initialize with a constant value.
random_normal(key, shape[, std])	Initialize with random normal values.
random_uniform(key, shape[, low, high])	Initialize with random uniform values.

Attributes Documentation

dtype

Return the dtype of the constant value.

shape

Return the shape of the constant value.

Methods Documentation

__call__(*args, **kwargs) → Array

Return the constant value, ignoring all inputs.

classmethod full(x: float = 0.0, shape: tuple[int, ...] = ())

Initialize with a constant value.

Parameters:

• x (float) – The constant value.

• shape (tuple[int, ...]) – The shape of the array.

Returns:

The initialized module.

Return type:

Constant

classmethod random_normal(key: Key[Array, ''] | UInt32[Array, '2'], shape: tuple[int, ...], std: float = 1.0)

Initialize with random normal values.

Parameters:

• key (PRNGKeyArray) – The key to use for random number generation.

• shape (tuple[int, ...]) – The shape of the array.

• std (float) – The standard deviation of the normal distribution.

Returns:

The initialized module.

Return type:

Constant

classmethod random_uniform(key: Key[Array, ''] | UInt32[Array, '2'], shape: tuple[int, ...], low: float = -1.0, high: float = 1.0)

Initialize with random uniform values.

Parameters:

• key (PRNGKeyArray) – The key to use for random number generation.

• shape (tuple[int, ...]) – The shape of the array.

• low (float) – The lower bound of the uniform distribution.

• high (float) – The upper bound of the uniform distribution.

Returns:

The initialized module.

Return type:

Constant