Skip to content

GitLab

  • Menu
Projects Groups Snippets
    • Loading...
  • Help
    • Help
    • Support
    • Community forum
    • Submit feedback
    • Contribute to GitLab
  • Sign in
  • wslda wslda
  • Project information
    • Project information
    • Activity
    • Labels
    • Members
  • Repository
    • Repository
    • Files
    • Commits
    • Branches
    • Tags
    • Contributors
    • Graph
    • Compare
  • Issues 0
    • Issues 0
    • List
    • Boards
    • Service Desk
    • Milestones
  • Merge requests 0
    • Merge requests 0
  • CI/CD
    • CI/CD
    • Pipelines
    • Jobs
    • Schedules
  • Deployments
    • Deployments
    • Environments
    • Releases
  • Monitor
    • Monitor
    • Incidents
  • Packages & Registries
    • Packages & Registries
    • Package Registry
    • Container Registry
    • Infrastructure Registry
  • Analytics
    • Analytics
    • CI/CD
    • Repository
    • Value stream
  • Wiki
    • Wiki
  • Snippets
    • Snippets
  • Activity
  • Graph
  • Create a new issue
  • Jobs
  • Commits
  • Issue Boards
Collapse sidebar
  • wtools
  • wsldawslda
  • Wiki
  • User defined parameters

User defined parameters · Changes

Page history
Create User defined parameters authored Dec 11, 2020 by Gabriel Wlazłowski's avatar Gabriel Wlazłowski
Show whitespace changes
Inline Side-by-side
User-defined-parameters.md 0 → 100644
View page @ cbe01261
# Custom parameters
The W-SLDA toolkit provides a framework for creating flexible parametrization of formulas. List of numbers (of double type) that can be passed from an input file to the code is called *custom parameters*. In the input file they start with tag:
```bash
# Custom parameters
params0 0.0 # value of the parameter
params1 1.0 # value of the parameter
params2 2.0 # value of the parameter
# ... and so on ...
```
Maximal number of parameters is specified in `predefines.h` by macro-variable:
```c
// Maximal number of parameters in params array
#define MAX_USER_PARAMS 32
```
When the program reads the input file custom parameters are uploaded into `params` array in a such way that *paramsk* is assigned to `params[k]`. Pointer to the array is passed to each user-defined function, for example:
```c
double v_ext(int ix, int iy, int iz, int it, int spin, double *params, size_t extra_data_size, void *extra_data)
{
// ADD HERE FORMULA FOR V_ext(r)
double V_ext = 0.0;
return V_ext;
}
```
During the self-iteration process, before `params` array is passed to user-defined functions the array can be processed by function:
```c
/**
* THIS FUNCTION IS CALLED DURING THE SELF-CONSISTENT PROCESS.
* After loading params array from input file, the parameters are processed by this routine.
* The routine is executed at beginning of each iteration.
* @param params array of size MAX_USER_PARAMS with parameters from input file.
* @param kF typical Fermi momentum scale of the problem.
* kF=referencekF if the referencekF tag is indicated in the input file,
* otherwise to kF value is assigned according formula kF=(3*pi^2*n)^{1/3}, where n corresponds to maximal density.
* You can also set kF at request in this function using (*kF)=myvalue;
* @param mu array with chemical potentials: mu[SPINA] and mu[SPINB].
* @param extra_data_size size of extra_data in bytes, if extra_data size=0 the optional data is not uploaded
* @param extra_data optional set of data uploaded by load_extra_data()
* */
void process_params(double *params, double *kF, double *mu, size_t extra_data_size, void *extra_data)
{
// PROCESS INPUT FILE PARAMETERS
}
```
Usage of `process_params( )` allows for parametrization of user-defined functions in terms of dimensionless values, which are typically more intuitive. For example:
```bash
params2 2.0 # barrier height, in units of chemical potential mu_a
```
and `process_params( )` converts *params2* into dimensional quantity
```c
void process_params(double *params, double *kF, double *mu, size_t extra_data_size, void *extra_data)
{
params[2]*=mu[SPINA]; // convert to dimensional form
}
```
# Passing external data to user-defined functions
To each user-defined function, arbitrary external data can be passed called *extra data*. The extra data is uploading before the self-consistent starts. In order to activate this option user must provide content of functions:
```c
/**
* This function provides size of extra_data array, in bytes.
* The extra_data of specified size will be allocated by the main process.
* This function is thread-safe.
* @param params with input file parameters.
* NOTE: the array contains bare input file values, not processed by process_params()!
* @return size of the extra_data array that needs to be allocated, if 0 then extra_data will not be allocated.
* */
size_t get_extra_data_size(double *params)
{
return 0;
}
```
and
```c
/**
* This function loads data into extra_data array.
* This function is thread-safe.
* @param size size of array computed using function get_extra_data_size()
* @param extra_data pointer to array that should be filled with data
* @param params with input file parameters.
* NOTE: the array contains bare input file values, not processed by process_params()!
* @return 0 if load is successful, otherwise return error code. If nonzero value is returned the main code will terminate.
* */
int load_extra_data(size_t size, void *extra_data, double *params)
{
return 0;
}
```
# Predefined variables
Each variable defined in `predefines.h` is accessible over entire file `problem-definition.h`. User can defined new variable in `predefines.h`, like
```c
#define MYVALUE 10.0;
```
which can be next used in user-defined functions:
```c
double v_ext(int ix, int iy, int iz, int it, int spin, double *params, size_t extra_data_size, void *extra_data)
{
// ADD HERE FORMULA FOR V_ext(r)
double V_ext = MYVALUE; // it will assign value as given in predefines.h
return V_ext;
}
```
\ No newline at end of file
Clone repository
  • API version
  • Automatic interpolations
  • Auxiliary tools
  • Browsing the code
  • Broyden algorithm
  • C and CUDA
  • Campaign of calculations
  • Checking correctness of settings
  • Chemical potentials control
  • Code & Results quality
  • Common failures of static codes
  • Common failures of time dependent codes
  • Computation domain
  • Configuring GPU machine
  • Constraining densities and potentials
View All Pages