|
CCL Library
1.0
|
This structure defines the workspace variables for solving the non-linear LM optimization. More...
#include <ccl_learn_ncl.h>
Data Fields | |
| int | dim_x |
| The dimensionality of the state variable. | |
| int | dim_n |
| The number of data points. | |
| int | dim_b |
| The number of basis functions. | |
| int | dim_y |
| The dimensionality of the action space. | |
| int | r_ok |
| check if the objective function is belowed the tolerence | |
| int | d_ok |
| check if the model parameters are belowed the tolerence | |
| double * | xc |
| A copy of the initial model parameters. | |
| double * | x |
| The flattened and updated model parameters. | |
| double * | xf |
| The finalised model parameters. | |
| double * | epsx |
| The tolerence of the model paramters. | |
| double | epsf |
| The tolerence fo the objective functions. | |
| double * | r |
| The returned value of the residual error. | |
| double * | J |
| The jacobian at x. | |
| double | S |
| The sqaure root if x is taken. | |
| double * | A |
| The dot product of J_T and J. | |
| double * | v |
| The dot product of J_T and r. | |
| double * | D |
| Automatic scaling. | |
| double | Rlo |
| The lower bound of R. | |
| double | Rhi |
| The upper bound of R. | |
| double | l |
| The adaptive learning rate. | |
| double | lc |
| Handling situation when learning rate happends to be 0. | |
| double * | d |
| The parameter improvement gradient. | |
| int | iter |
| The iteration number. | |
| double * | xd |
| The next x. | |
| double * | rd |
| The residual error at xd. | |
| double | Sd |
| The squared error if xd is taken. | |
| double | dS |
| The denomitor of the sqaured error is xd is taken. | |
| double | R |
| The reduction if xd is taken. | |
| double | nu |
| The coefficient of changing learning rate. | |
| double * | d_T |
| The matrix transpose of d. | |
| double * | J_T |
| The matrix transpose of J. | |
| double * | tmp |
| The temporal variable. | |
| double * | rd_T |
| The matrix transpose of rd. | |
| gsl_matrix * | D_pinv |
| The peudo-inverse of D. | |
| gsl_vector * | A_d |
| The A matrix at xd. | |
| gsl_matrix * | A_inv |
| The peudo-inverse of A. | |
| gsl_vector * | A_inv_diag |
| The vector of diagonal elements of A matrix. | |
| gsl_vector * | r_T |
| The vector transpose of r. | |
This structure defines the workspace variables for solving the non-linear LM optimization.
This structure containts the memory of the workspace variables for sovling the non-linear LM optimization problem. It should always initialise with ccl_learn_model_ws_alloc and destroy with ccl_learn_model_ws_free.
1.8.6