reconstruction Module Reference

Defines the base class of the type reconstruction_class. This class contains the minimization algorithm. More...

Data Types
type	reconstruction_class
	Base class containing all the data needed to reconstruct a model. More...
interface	reconstruction_write_step
	Interface for writing out step data. More...

Functions/Subroutines
type(reconstruction_class) function, pointer	reconstruction_construct (num_steps, num_signals, num_derived_parameters, num_parameters, step_type, step_max, cut_svd, cut_eff, cut_marg_eff, cut_delta_a, cut_dg2, last_para_signal, cut_inv_svd, use_central)
	Construct a reconstruction_class object. More...
subroutine	reconstruction_destruct (this)
	Deconstruct a reconstruction_class object. More...
integer function	reconstruction_get_k_use (this, num_sv)
	Get the number of singular values to use. More...
real(rprec) function	reconstruction_get_exp_dg2 (this, delta_a)
	Get the expected dg^2 size. More...
real(rprec) function	reconstruction_get_exp_g2 (this, delta_a)
	Get the expected g^2 size. More...
real(rprec) function	reconstruction_get_g2 (this)
	Get the current g^2. More...
real(rprec) function	reconstruction_get_lastg2 (this)
	Get the last g^2. More...
real(rprec) function	reconstruction_get_dg2 (this)
	Get the change in g^2. More...
logical function	reconstruction_eval_e (this, signals, a_model, gaussp, eq_steps, iou, eq_comm)
	Evaluate the e vector for a reconstruction step. More...
subroutine	reconstruction_eval_f (this, derived_params, a_model)
	Evaluate the f vector for a reconstruction step. More...
subroutine	reconstruction_eval_jacobians (this, signals, derived_params, locks, a_model, gaussp, params, eq_steps, iou, recon_comm, eq_comm)
	Evaluate the jacobian. More...
integer function	reconstruction_eval_step (this, signals, derived_params, locks, a_model, gaussp, params, eq_steps, iou, recon_comm, eq_comm)
	Evaluate the proper quasi-Newton step. More...
real(rprec) function	reconstruction_sl_step (this, k_use, step_size, delta_a)
	Take a straight line step. More...
real(rprec) function	reconstruction_seg_step (this, k_use, step_size, delta_a)
	Take a segmented path step. More...
real(rprec) function	reconstruction_lm_step (this, k_use, step_size, delta_a)
	Take a Levenberg-Marquardt step. More...
real(rprec) function	reconstruction_lm_rootfind (this, k_use, lambda_default, step_size, utdote)
	Root find for Levenberg-Marquardt step. More...
real(rprec) function	reconstruction_lm_function (this, f_sqrd, step_size_sqrd, lambda)
	Levenberg-Marquardt function. More...
logical function	reconstruction_step (this, signals, derived_params, locks, a_model, gaussp, params, eq_steps, iou, recon_comm, eq_comm)
	Step the reconstruction. More...
logical function	reconstruction_try_step (this, signals, derived_params, locks, a_model, gaussp, params, eq_steps, max_step, iou, recon_comm, eq_comm)
	Try step size. More...
subroutine	reconstruction_eval_sem (this, params, signals, derived_params)
	Evaluate the parameter covariance and signal effectiveness. More...
subroutine	reconstruction_invert_matrix (this, matrix, sub, name)
	Invert an M x N matrix. More...
subroutine	reconstruction_normalize_correlations (params)
	Normalize a correlation matrix. More...
subroutine	reconstruction_write (this, iou)
	Write out a reconstruction. More...
subroutine	reconstruction_write_step1 (this, step_type, iou)
	Write out a reconstruction step attempt. More...
subroutine	reconstruction_write_step2 (this, iou)
	Write out a final reconstruction step. More...
subroutine	reconstruction_restart (this, result_ncid, current_step, signals, derived_params, a_model)
	Restart from result file. More...
subroutine	reconstruction_sync_state (this, recon_comm)
	Syncronize state to children. More...
subroutine	reconstruction_sync_svd (this, recon_comm)
	Syncronize svd. More...
subroutine	reconstruction_sync_parent (this, params, num_signal, num_derived_params, stride, offset, recon_comm)
	Syncronize child state back to parent. More...

Variables
integer, parameter	reconstruction_no_step_type = -1
	No reconstruction step type.
integer, parameter	reconstruction_sl_step_type = 0
	Straight line reconstruction step type.
integer, parameter	reconstruction_lm_step_type = 1
	Levenberg-Marquardt reconstruction step type.
integer, parameter	reconstruction_seg_step_type = 2
	Segmented path reconstruction step type.
integer, parameter	reconstruction_max_step_try = 5
	Maximum number of reconstruction step attemps.

Detailed Description

Defines the base class of the type reconstruction_class. This class contains the minimization algorithm.

Function/Subroutine Documentation

reconstruction_construct()

type (reconstruction_class) function, pointer reconstruction::reconstruction_construct	(	integer, intent(in)	num_steps,
		integer, intent(in)	num_signals,
		integer, intent(in)	num_derived_parameters,
		integer, intent(in)	num_parameters,
		character (len=*), intent(in)	step_type,
		real (rprec), intent(in)	step_max,
		real (rprec), intent(in)	cut_svd,
		real (rprec), intent(in)	cut_eff,
		real (rprec), intent(in)	cut_marg_eff,
		real (rprec), intent(in)	cut_delta_a,
		real (rprec), intent(in)	cut_dg2,
		integer, intent(in)	last_para_signal,
		real (rprec), intent(in)	cut_inv_svd,
		logical, intent(in)	use_central
	)

Construct a reconstruction_class object.

Allocates memory and initializes a reconstruction_class object.

Parameters

[in]	num_steps	Number of reconstruction steps. v3fit_input::nrstep
[in]	num_signals	Number of signals.
[in]	num_derived_parameters	Number of derived parameters.
[in]	num_parameters	Number of reconstruction parameters.
[in]	step_type	Reconstruction step type. v3fit_input::step_type
[in]	step_max	Reconstruction step size. v3fit_input::astep_max
[in]	cut_svd	Cutoff value for relative singular values. v3fit_input::cut_svd
[in]	cut_eff	Cutoff value for expected step efficiency. v3fit_input::cut_eff
[in]	cut_marg_eff	Cutoff value for expected marginal step efficiency. v3fit_input::cut_marg_eff
[in]	cut_delta_a	Cutoff value for expected step size. v3fit_input::cut_delta_a
[in]	cut_dg2	Cutoff value for expected change in g^2. v3fit_input::cut_dg2
[in]	last_para_signal	Index of the last signal that can be parallelized.
[in]	cut_inv_svd	Cutoff value for the pseudo inverse singular values.

Returns

A pointer to a constructed reconstruction_class object.

Definition at line 174 of file reconstruction.f.

reconstruction_destruct()

subroutine reconstruction::reconstruction_destruct

(

type (reconstruction_class), pointer

this

)

Deconstruct a reconstruction_class object.

Deallocates memory and uninitializes a reconstruction_class object.

Parameters

[in,out]

this

A reconstruction_class instance.

Definition at line 280 of file reconstruction.f.

reconstruction_eval_e()

logical function reconstruction::reconstruction_eval_e	(	type (reconstruction_class), intent(inout)	this,
		type (signal_pointer), dimension(:), intent(inout)	signals,
		type (model_class), pointer	a_model,
		type (gaussp_class_pointer), dimension(:), intent(inout)	gaussp,
		integer, intent(inout)	eq_steps,
		integer, intent(in)	iou,
		integer, intent(in)	eq_comm
	)

Evaluate the e vector for a reconstruction step.

Evaluates the e vector after soving the equilibrium.

See also

signal::signal_get_e

Parameters

[in,out]	this	A reconstruction_class instance.
[in,out]	signals	Array of signal objects.
[in,out]	a_model	A model instance.
[in,out]	gaussp	Array of guassian_process::gaussp_class objects.
[in,out]	eq_steps	Number of steps for the equilibrium to take.
[in]	iou	Input/output file to write log messages to.
[in]	eq_comm	MPI communicator for the child equilibrium processes.

Returns

True if the equilibrium converged and false if otherwise.

Definition at line 670 of file reconstruction.f.

reconstruction_eval_f()

subroutine reconstruction::reconstruction_eval_f	(	type (reconstruction_class), intent(inout)	this,
		type (param_pointer), dimension(:), intent(in)	derived_params,
		type (model_class), intent(inout)	a_model
	)

Evaluate the f vector for a reconstruction step.

Evaluates the f vector after soving the equilibrium.

See also

param::param_get_value

Parameters

[in,out]	this	A reconstruction_class instance.
[in]	derived_params	Array of param objects.
[in,out]	a_model	A model instance.

Definition at line 738 of file reconstruction.f.

reconstruction_eval_jacobians()

subroutine reconstruction::reconstruction_eval_jacobians	(	type (reconstruction_class), intent(inout)	this,
		type (signal_pointer), dimension(:), intent(inout)	signals,
		type (param_pointer), dimension(:), intent(in)	derived_params,
		type (param_pointer), dimension(:), intent(in)	locks,
		type (model_class), pointer	a_model,
		type (gaussp_class_pointer), dimension(:), intent(inout)	gaussp,
		type (param_pointer), dimension(:), intent(inout)	params,
		integer, intent(inout)	eq_steps,
		integer, intent(in)	iou,
		integer, intent(in)	recon_comm,
		integer, intent(in)	eq_comm
	)

Evaluate the jacobian.

Evaluates the jacobian. When using MPI, reach row of the jacobian is evaluated in parallel.

Parameters

[in,out]	this	A reconstruction_class instance.
[in,out]	signals	Array of signal objects.
[in]	derived_params	Array of derived param objects.
[in,out]	a_model	A model instance.
[in,out]	params	Array of parameter objects.
[in,out]	eq_steps	Number of steps for the equilibrium to take.
[in]	iou	Input/output file to write log messages to.
[in]	recon_comm	MPI communicator for the child jacobian processes.
[in]	eq_comm	MPI communicator for the child equilibrium processes.

Definition at line 785 of file reconstruction.f.

reconstruction_eval_sem()

subroutine reconstruction::reconstruction_eval_sem	(	type (reconstruction_class), intent(inout)	this,
		type (param_pointer), dimension(:), intent(inout)	params,
		type (signal_pointer), dimension(:), intent(inout)	signals,
		type (param_pointer), dimension(:), intent(inout)	derived_params
	)

Evaluate the parameter covariance and signal effectiveness.

Calculates the signal effectiveness matrix. This value is stored in parameter::param_class::sem. To find the signal effectiveness, an intermediate step computes the parameter covariance. Computes the derived parameter covariance matrix also.

Parameters

[in,out]	this	A reconstruction_class instance.
[in,out]	params	Array of parameter objects.
[in,out]	signals	Array of signal objects.
[in,out]	derived_params	Array of parameter objects.

Definition at line 2137 of file reconstruction.f.

reconstruction_eval_step()

integer function reconstruction::reconstruction_eval_step	(	type (reconstruction_class), intent(inout)	this,
		type (signal_pointer), dimension(:), intent(inout)	signals,
		type (param_pointer), dimension(:), intent(in)	derived_params,
		type (param_pointer), dimension(:), intent(in)	locks,
		type (model_class), pointer	a_model,
		type (gaussp_class_pointer), dimension(:), intent(inout)	gaussp,
		type (param_pointer), dimension(:), intent(inout)	params,
		integer, intent(inout)	eq_steps,
		integer, intent(in)	iou,
		integer, intent(in)	recon_comm,
		integer, intent(in)	eq_comm
	)

Evaluate the proper quasi-Newton step.

Evaluates the jacobian, performs the SVD and computes the quasi-Newton step.

Parameters

[in,out]	this	A reconstruction_class instance.
[in,out]	signals	Array of signal objects.
[in]	derived_params	Array of derived param objects.
[in,out]	a_model	A model instance.
[in,out]	params	Array of parameter objects.
[in,out]	eq_steps	Number of steps for the equilibrium to take.
[in]	iou	Input/output file to write log messages to.
[in]	recon_comm	MPI communicator for the child jacobian processes.
[in]	eq_comm	MPI communicator for the child equilibrium processes.

Returns

The number of singular values to use.

Definition at line 1098 of file reconstruction.f.

reconstruction_get_dg2()

real (rprec) function reconstruction::reconstruction_get_dg2

(

type (reconstruction_class), intent(in)

this

)

Get the change in g^2.

Gets the change in g^2 from the last reconstruction step to the current step.

Parameters

[in]

this

A reconstruction_class instance.

Returns

The last g^2.

Definition at line 631 of file reconstruction.f.

reconstruction_get_exp_dg2()

real (rprec) function reconstruction::reconstruction_get_exp_dg2	(	type (reconstruction_class), intent(in)	this,
		real (rprec), dimension(:), intent(in)	delta_a
	)

Get the expected dg^2 size.

Estimates the change in g^2. Equation 22 in Hanson et. al. doi:10.1088/0029-5515/49/7/075031 Note: there is a typo in equation 22. The - sign in the last term should be a plus sign.

Parameters

[in]	this	A reconstruction_class instance.
[in]	delta_a	The normalized parameter step.

Returns

Change expected change in g^2.

Definition at line 496 of file reconstruction.f.

reconstruction_get_exp_g2()

real (rprec) function reconstruction::reconstruction_get_exp_g2	(	type (reconstruction_class), intent(in)	this,
		real (rprec), dimension(:), intent(in)	delta_a
	)

Get the expected g^2 size.

Estimates the g^2. Equation 22 in Hanson et. al. doi:10.1088/0029-5515/49/7/075031

Parameters

[in]	this	A reconstruction_class instance.
[in]	delta_a	The normalized parameter step.

Returns

Expected g^2.

Definition at line 538 of file reconstruction.f.

reconstruction_get_g2()

real (rprec) function reconstruction::reconstruction_get_g2

(

type (reconstruction_class), intent(in)

this

)

Get the current g^2.

Gets the g^2 value for the current reconstruction step.

Parameters

[in]

this

A reconstruction_class instance.

Returns

Current g^2.

Definition at line 570 of file reconstruction.f.

reconstruction_get_k_use()

integer function reconstruction::reconstruction_get_k_use	(	type (reconstruction_class), intent(inout)	this,
		integer, intent(in)	num_sv
	)

Get the number of singular values to use.

Number of sigular values to uses is determined by the various cutoffs.

Parameters

[in,out]	this	A reconstruction_class instance.
[in]	num_sv	Total number of sigular values.

Returns

The number of sigular values to use.

Definition at line 399 of file reconstruction.f.

reconstruction_get_lastg2()

real (rprec) function reconstruction::reconstruction_get_lastg2

(

type (reconstruction_class), intent(in)

this

)

Get the last g^2.

Gets the g^2 value for the last reconstruction step.

Parameters

[in]

this

A reconstruction_class instance.

Returns

The last g^2.

Definition at line 600 of file reconstruction.f.

reconstruction_invert_matrix()

subroutine reconstruction::reconstruction_invert_matrix	(	type (reconstruction_class), intent(inout)	this,
		real (rprec), dimension(:,:), pointer	matrix,
		character (len=*), intent(in)	sub,
		character (len=*), intent(in)	name
	)

Invert an M x N matrix.

Perform a pseudo-inversion using a singular value decomposition. This subroutine destroys the original matrix.

Parameters

[in,out]	this	A reconstruction_class instance.
[in,out]	matrix	The matrix to invert.
[in]	sub	The subroutine of the matrix this was called from.
[in]	name	The name of the matrix to be inverted.

Definition at line 2256 of file reconstruction.f.

reconstruction_lm_function()

real (rprec) function reconstruction::reconstruction_lm_function	(	type (reconstruction_class), intent(inout)	this,
		real (rprec), dimension(:), intent(in)	f_sqrd,
		real (rprec), intent(in)	step_size_sqrd,
		real (rprec), intent(in)	lambda
	)

Levenberg-Marquardt function.

Function used by reconstruction_lm_rootfind for a Levenberg-Marquardt step.

Parameters

[in,out]	this	A reconstruction_class instance.
[in]	f_sqrd	f^2
[in]	step_size_sqrd	Square of the current step size.
[in]	lambda	Lambda value.

Returns

F(lambda)

Definition at line 1592 of file reconstruction.f.

reconstruction_lm_rootfind()

real (rprec) function reconstruction::reconstruction_lm_rootfind	(	type (reconstruction_class), intent(inout)	this,
		integer, intent(in)	k_use,
		real (rprec), intent(in)	lambda_default,
		real (rprec), intent(in)	step_size,
		real (rprec), dimension(:), intent(in)	utdote
	)

Root find for Levenberg-Marquardt step.

Find the root of the reconstruction_lm_function for a Levenberg-Marquardt step.

Parameters

[in,out]	this	A reconstruction_class instance.
[in]	k_use	Number of singular values to use.
[in]	lambda_default	The default lambda.
[in]	step_size	Current step size limit.
[in]	utdote	Vector of U^T * e

Returns

The root lambda.

Definition at line 1439 of file reconstruction.f.

reconstruction_lm_step()

real (rprec) function reconstruction::reconstruction_lm_step	(	type (reconstruction_class), intent(inout)	this,
		integer, intent(in)	k_use,
		real (rprec), intent(in)	step_size,
		real (rprec), dimension(:), intent(out)	delta_a
	)

Take a Levenberg-Marquardt step.

Determines the normalized change in parameters for a Levenberg-Marquardt step.

Parameters

[in,out]	this	A reconstruction_class instance.
[in]	k_use	Number of singular values to use.
[in]	step_size	Current step size limit.
[out]	delta_a	Normalized change in parameter.

Returns

The step size used.

Definition at line 1368 of file reconstruction.f.

reconstruction_normalize_correlations()

subroutine reconstruction::reconstruction_normalize_correlations

(

type (param_pointer), dimension(:), intent(inout)

params

)

Normalize a correlation matrix.

Normalizes a correlation matrix by dividing by the sigmas for that element. This method is static so it doesn't require a this parameter.

Parameters

[in,out]

params

Array of parameter to normalize the correlations of.

Definition at line 2335 of file reconstruction.f.

reconstruction_restart()

subroutine reconstruction::reconstruction_restart	(	type (reconstruction_class), intent(inout)	this,
		integer, intent(in)	result_ncid,
		integer, intent(in)	current_step,
		type (signal_pointer), dimension(:), intent(inout)	signals,
		type (param_pointer), dimension(:), intent(in)	derived_params,
		type (model_class), pointer	a_model
	)

Restart from result file.

Restarts the reconstruction from the result file. Note, that there is an external step counter and an internal step counter. Internally we will restart current step at zero.

Parameters

[in,out]	this	A reconstruction_class instance.
[in]	result_ncid	NetCDF file id of the result file.
[in]	current_step	Step index to read variables from.
[in,out]	a_model	A model instance.

Definition at line 2555 of file reconstruction.f.

reconstruction_seg_step()

real (rprec) function reconstruction::reconstruction_seg_step	(	type (reconstruction_class), intent(inout)	this,
		integer, intent(in)	k_use,
		real (rprec), intent(in)	step_size,
		real (rprec), dimension(:), intent(out)	delta_a
	)

Take a segmented path step.

Determines the normalized change in parameters for a segmented path step.

Parameters

[in,out]	this	A reconstruction_class instance.
[in]	k_use	Number of singular values to use.
[in]	step_size	Current step size limit.
[out]	delta_a	Normalized change in parameter.

Returns

The step size used.

Definition at line 1278 of file reconstruction.f.

reconstruction_sl_step()

real (rprec) function reconstruction::reconstruction_sl_step	(	type (reconstruction_class), intent(inout)	this,
		integer, intent(in)	k_use,
		real (rprec), intent(in)	step_size,
		real (rprec), dimension(:), intent(out)	delta_a
	)

Take a straight line step.

Determines the normalized change in parameters for a straight line step.

Parameters

[in,out]	this	A reconstruction_class instance.
[in]	k_use	Number of singular values to use.
[in]	step_size	Current step size limit.
[out]	delta_a	Normalized change in parameter.

Returns

The step size used.

Definition at line 1236 of file reconstruction.f.

reconstruction_step()

logical function reconstruction::reconstruction_step	(	type (reconstruction_class), intent(inout)	this,
		type (signal_pointer), dimension(:), intent(inout)	signals,
		type (param_pointer), dimension(:), intent(in)	derived_params,
		type (param_pointer), dimension(:), intent(in)	locks,
		type (model_class), pointer	a_model,
		type (gaussp_class_pointer), dimension(:), intent(inout)	gaussp,
		type (param_pointer), dimension(:), intent(inout)	params,
		integer, intent(inout)	eq_steps,
		integer, intent(in)	iou,
		integer, intent(in)	recon_comm,
		integer, intent(in)	eq_comm
	)

Step the reconstruction.

Advance the reconstruction by a single step. This is the main V3FIT algorithm.

Parameters

[in,out]	this	A reconstruction_class instance.
[in,out]	signals	Array of signal objects.
[in,out]	locks	Array of lock objects.
[in,out]	a_model	A model instance.
[in,out]	gaussp	Array of guassian_process::gaussp_class objects.
[in,out]	params	Array of v3fit_params::param_class objects.
[in,out]	eq_steps	Number of steps the equilibrium took.
[in]	iou	Input/output file to write log messages to.
[in]	recon_comm	MPI communicator for the child jacobian processes.
[in]	eq_comm	MPI communicator for the child equilibrium processes.

Returns

True if the step was successfull and false otherwise.

Definition at line 1647 of file reconstruction.f.

reconstruction_sync_parent()

subroutine reconstruction::reconstruction_sync_parent	(	type (reconstruction_class), intent(inout)	this,
		type (param_pointer), dimension(:), intent(inout)	params,
		integer, intent(in)	num_signal,
		integer, intent(in)	num_derived_params,
		integer, dimension(:), intent(in)	stride,
		integer, dimension(:), intent(in)	offset,
		integer, intent(in)	recon_comm
	)

Syncronize child state back to parent.

Syncs data between the parent and child processes. If MPI support is not compiled in this subroutine reduces to a no op.

Parameters

[in,out]	this	A reconstruction_class instance.
[in]	num_params	Number of reconstruction parameters.
[in]	num_signal	Number of signals.
[in]	num_derived_params	Number of derived parameters.
[in]	stride	Length work size for each process.
[in]	offset	Start offset for each process.
[in]	recon_comm	A MPI recon_comm handle.

Definition at line 2720 of file reconstruction.f.

reconstruction_sync_state()

subroutine reconstruction::reconstruction_sync_state	(	type (reconstruction_class), intent(inout)	this,
		integer, intent(in)	recon_comm
	)

Syncronize state to children.

Syncs data between the parent and child processes. If MPI support is not compiled in this subroutine reduces to a no op.

Parameters

[in,out]	this	A reconstruction_class instance.
[in]	recon_comm	A MPI recon_comm handle.

Definition at line 2619 of file reconstruction.f.

reconstruction_sync_svd()

subroutine reconstruction::reconstruction_sync_svd	(	type (reconstruction_class), intent(inout)	this,
		integer, intent(in)	recon_comm
	)

Syncronize svd.

Syncs the svd information to the child processes so these processes can sample parameter space for the step sizes.

Parameters

[in,out]	this	A reconstruction_class instance.
[in]	recon_comm	A MPI recon_comm handle.

Definition at line 2660 of file reconstruction.f.

reconstruction_try_step()

logical function reconstruction::reconstruction_try_step	(	type (reconstruction_class), intent(inout)	this,
		type (signal_pointer), dimension(:), intent(inout)	signals,
		type (param_pointer), dimension(:), intent(in)	derived_params,
		type (param_pointer), dimension(:), intent(in)	locks,
		type (model_class), pointer	a_model,
		type (gaussp_class_pointer), dimension(:), intent(inout)	gaussp,
		type (param_pointer), dimension(:), intent(inout)	params,
		integer, intent(inout)	eq_steps,
		real (rprec), intent(inout)	max_step,
		integer, intent(in)	iou,
		integer, intent(in)	recon_comm,
		integer, intent(in)	eq_comm
	)

Try step size.

Sample reconstruction by trying to take a specified step size.

Parameters

[in,out]	this	A reconstruction_class instance.
[in,out]	signals	Array of signal objects.
[in,out]	locks	Array of lock objects.
[in,out]	a_model	A model instance.
[in,out]	params	Array of parameter objects.
[in,out]	eq_steps	Number of steps for the equilibrium to take.
[in]	max_step	Upper limit of the step size to attemp.
[in]	iou	Input/output file to write log messages to.
[in]	recon_comm	MPI communicator for the child jacobian processes.
[in]	eq_comm	MPI communicator for the child equilibrium processes.
[out]	param_value	Parameter corresponding to the smallest g^2 value.

Returns

The lowest q^2 value.

Definition at line 1810 of file reconstruction.f.

reconstruction_write()

subroutine reconstruction::reconstruction_write	(	type (reconstruction_class), pointer	this,
		integer, intent(in)	iou
	)

Write out a reconstruction.

Write out the final information about a reconstruction. This writes out a table of the g^2, expected g^2, number of sigular values used and the normalized step size used.

Parameters

[in]	this	A reconstruction_class instance.
[in]	iou	Input/output file to write final result to.

Definition at line 2373 of file reconstruction.f.

reconstruction_write_step1()

subroutine reconstruction::reconstruction_write_step1	(	type (reconstruction_class), intent(in)	this,
		character (len=*), intent(in)	step_type,
		integer, intent(in)	iou
	)

Write out a reconstruction step attempt.

Write out information about a reconstruction step. This writes the number of singular values use, the step type, the normalized step size and the expected g^2.

Parameters

[in]	this	A reconstruction_class instance.
[in]	step_type	The step type used.
[in]	iou	Input/output file to write final result to.

Definition at line 2427 of file reconstruction.f.

reconstruction_write_step2()

subroutine reconstruction::reconstruction_write_step2	(	type (reconstruction_class), intent(in)	this,
		integer, intent(in)	iou
	)

Write out a final reconstruction step.

Write out information about a reconstruction step. This writes the number of singular values use, the step type, the normalized step size and the expected g^2.

Parameters

[in]	this	A reconstruction_class instance.
[in]	iou	Input/output file to write final result to.

Definition at line 2471 of file reconstruction.f.