Transforms

transforms ¶

Transform functions for graph-based simulators.

Functions¶

beta_spin_magnitude ¶

beta_spin_magnitude(*, alpha: float, beta: float, key: Array, reference: Array | None = None, n_samples: int | None = None, minimum: float = 0.0, maximum: float = 1.0) -> Array

Draw spin magnitudes from a rescaled Beta distribution.

constant_like ¶

constant_like(reference: Array, value: float) -> Array

Return a constant array with the same shape as reference.

Parameters:

Name	Type	Description	Default
`reference`	`Array`	Array providing the target shape.	required
`value`	`float`	Scalar value to broadcast across the output.	required

Returns:

Type	Description
`Array`	An array with the same shape as `reference` filled with `value`.

gaussian_chi_eff ¶

gaussian_chi_eff(*, key: Array | None = None, reference: Array | None = None, n_samples: int | None = None, mean: float = 0.0, sigma: float = 0.1, minimum: float = -1.0, maximum: float = 1.0, chi_eff: Array | None = None, component: str | None = None, mass_1: Array | None = None, mass_2: Array | None = None, spin_magnitude_1: Array | None = None, spin_magnitude_2: Array | None = None) -> Array

Draw or project an effective aligned-spin distribution.

When component is omitted, this returns chi_eff samples drawn from a clipped Gaussian. When component is "primary" or "secondary", the function projects the supplied or sampled chi_eff values onto the requested spin-z component while respecting the available spin magnitudes.

identity ¶

identity(*, value: Array) -> Array

Return value unchanged.

Useful for aliasing one parameter to another in a graph config (e.g. setting theta_jn equal to inclination for aligned-spin populations).

Parameters:

Name	Type	Description	Default
`value`	`Array`	Array-like value to pass through.	required

Returns:

Type	Description
`Array`	`value` as a JAX array.

isotropic_declination ¶

isotropic_declination(*, key: Array, reference: Array | None = None, n_samples: int | None = None) -> Array

Draw declinations with sin(declination) uniform on [-1, 1].

Produces an isotropic sky distribution in declination, i.e. angles in [-pi/2, pi/2] with arcsin of a uniform variate.

Parameters:

Name	Type	Description	Default
`key`	`Array`	JAX PRNG key.	required
`reference`	`Array \| None`	Optional array providing the output shape.	`None`
`n_samples`	`int \| None`	Number of samples, used when `reference` is omitted.	`None`

Returns:

Type	Description
`Array`	Declination angles in radians on `[-pi/2, pi/2]`.

isotropic_spin_orientation ¶

isotropic_spin_orientation(*, key: Array, reference: Array | None = None, n_samples: int | None = None) -> Array

Draw polar angles with cos(tilt) uniform on [-1, 1].

luminosity_distance_to_redshift ¶

luminosity_distance_to_redshift(luminosity_distance: Array, *, hubble_constant: float = PLANCK18_H0_KM_S_MPC, omega_m: float = PLANCK18_OMEGA_M, max_redshift: float = DEFAULT_MAX_REDSHIFT, n_grid: int = DEFAULT_LOOKUP_GRID_SIZE) -> Array

Convert luminosity distance to redshift with a flat-ΛCDM lookup.

Parameters:

Name	Type	Description	Default
`luminosity_distance`	`Array`	Luminosity distance in Mpc.	required
`hubble_constant`	`float`	Hubble constant in km / s / Mpc.	`PLANCK18_H0_KM_S_MPC`
`omega_m`	`float`	Matter density.	`PLANCK18_OMEGA_M`
`max_redshift`	`float`	Largest redshift supported by the lookup table.	`DEFAULT_MAX_REDSHIFT`
`n_grid`	`int`	Number of tabulation points.	`DEFAULT_LOOKUP_GRID_SIZE`

Returns:

Type	Description
`Array`	Redshift inferred from `luminosity_distance`.

maximum ¶

maximum(left: Array, right: Array) -> Array

Return the elementwise maximum of two array-like inputs.

minimum ¶

minimum(left: Array, right: Array) -> Array

Return the elementwise minimum of two array-like inputs.

multiply ¶

multiply(left: Array, right: Array) -> Array

Multiply two array-like inputs elementwise.

Parameters:

Name	Type	Description	Default
`left`	`Array`	Left multiplicand.	required
`right`	`Array`	Right multiplicand.	required

Returns:

Type	Description
`Array`	Elementwise product of `left` and `right`.

spherical_to_cartesian_x ¶

spherical_to_cartesian_x(*, magnitude: Array, tilt: Array, azimuth: Array) -> Array

Project spherical coordinates onto the Cartesian x component.

spherical_to_cartesian_y ¶

spherical_to_cartesian_y(*, magnitude: Array, tilt: Array, azimuth: Array) -> Array

Project spherical coordinates onto the Cartesian y component.

spherical_to_cartesian_z ¶

spherical_to_cartesian_z(*, magnitude: Array, tilt: Array) -> Array

Project spherical coordinates onto the Cartesian z component.

take_row ¶

take_row(*, matrix: Array, index: int) -> Array

Select one row from a 2-D array.

Used to split a jointly sampled (k, n_samples) array (e.g. a component mass pair) into individual 1-D parameter columns.

Parameters:

Name	Type	Description	Default
`matrix`	`Array`	Array whose leading axis is indexed.	required
`index`	`int`	Row index to select.	required

Returns:

Type	Description
`Array`	The selected row as a 1-D array.