smica.instrument.LogLinearBeam Class Reference

Inheritance diagram for smica.instrument.LogLinearBeam:

Public Member Functions
def	transfert_function
def	get_thetascore
def	get_thetaroot
def	get_thetafim
def	apply_beam
Public Member Functions inherited from smica.instrument.Instrument
def	__init__
def	get_dim
def	get_theta
def	set_theta
def	pre_alphascore
def	transfert_function
def	apply_beam

Additional Inherited Members
Public Attributes inherited from smica.instrument.Instrument
	ndet
	nvec
	nbin
	ndim
	vectors
	is this actually used ?
	alpha
	binning

Detailed Description

Implementation of log-linear observation model

The log-linear model writes

.. math:: b_nl = \exp(\sum_k \alpha_nk \Delta_nkl)

where n is the detector index, l the bin index and k the basis
index. The matrix observation model B is diagonal with b_nl as
element.

Member Function Documentation

def smica.instrument.LogLinearBeam.apply_beam	(		self,
			cov_source,
			bin = None
	)

Return the application of the transfert function to the
covariance sources ie B^dag S B for all bin

def smica.instrument.LogLinearBeam.get_thetafim	(		self,
			iRr,
			nmode,
			signal_comp,
			brute_force = False
	)

Return Fisher Information Matrix.

Parameters
----------
iRr : array-like
  shape of (ndet, ndet, nbin). iRr is the square root of the
  inverse of the sum of the 'sky component' covariance

signal_components : array-like, shape (ndet, ndet, nbin)
  The sum of signal components

nmode : ndarray of int
  The number of modes for each bin of shape (1, nbin).

Returns
----------
array-like, shape (N, N), where N is the number of free parameters.

def smica.instrument.LogLinearBeam.get_thetaroot	(		self,
			iRr,
			signal_component,
			bin = 0
	)

Return root contribution to the Fisher Information Matrix
for a given bin.

Parameters
----------
iRr : array-like, shape (ndet, ndet).
iRr is the square root of the inverse of the covariance of the
model, which is constant for all components at bin bin

signal_components : array-like, shape (ndet, ndet)
  The sum of signal component at bin bin. Beam must not be applied.

bin : int, bin number

Returns
----------
  array-like, shape (ndet * ndet, ndim) where ndim is the number
of parameters in the order [vec1-det1 vec1-det2 ... vec2-det1
...]. out[0, :] should match beam.get_theta(). A reshape of
out[0, :].reshape((nvec, ndet)) should match the output of
beam.theta.

def smica.instrument.LogLinearBeam.get_thetascore	(		self,
			D,
			observed_cov
	)

Return first derivative as a column vector.

Parameters
----------
D : ndarray
  A precomputed ndarray of shape (ndet, ndet, nbin). SMICA wizard
  know what is it.

observed_covariance : ndarray
  A ndarray of shape (ndet, ndet, nbin) corresponding to the
  sum of observed component covariance.

def smica.instrument.LogLinearBeam.transfert_function	(		self,
			bin = None
	)

Return transfert function as a (ndet, ndet, nbin)
array. bin can be specified.

---

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

• smica/instrument.py