sam::FalsifyingData::Parameters¶
Module: Parameters of Hacking Strategies
#include <HackingStrategy.h>
Public Attributes¶
Name | |
---|---|
HackingMethod | name |
std::string | approach |
std::string | switching_direction |
std::string | selection_method |
HackingTarget | target |
int | n_attempts Number of trials. |
size_t | num Number of observations to be perturbed. |
std::optional< UnivariateDistribution > | noise_dist Distribution of noise. |
std::vector< std::string > | stopping_cond_defs Stopping condition PolicyChain definitions. |
std::optional< float > | defensibility The defensibility factor of the strategy. |
std::optional< float > | prevalence The prevalence factor of the strategy. |
HackingStage | stage |
Detailed Description¶
struct sam::FalsifyingData::Parameters;
Falsifying Data Parameters
Example usage:
{
"name": "FalsifyingData",
}
Filename: .json
Public Attributes Documentation¶
variable name¶
HackingMethod name = HackingMethod::FalsifyingData;
variable approach¶
std::string approach {"perturbation"};
Falsification approach. We've discussed two possible way of doing this
- perturbation, perturbing a value
- group swapping, swapping values between groups
- group switching, moving values between groups
variable switching_direction¶
std::string switching_direction {"control-to-treatment"};
Switching direction
- control-to-treatment
- treatment-to-control
variable selection_method¶
std::string selection_method {"random"};
Swapping Method
- random
- smart
variable target¶
HackingTarget target {HackingTarget::Both};
Indicates which outcome variables are going to be targeted,
- control
- treatment
- both
variable n_attempts¶
int n_attempts {1};
Number of trials.
Todo: To be implemented
Indicates a set of rule that is going to be used to select the target group
variable num¶
size_t num;
Number of observations to be perturbed.
variable noise_dist¶
std::optional< UnivariateDistribution > noise_dist = [makeUnivariateDistribution](/doxygen/Files/_distributions_8h/#function-makeunivariatedistribution)({
{"dist", "normal_distribution"},
{"mean", 0},
{"stddev", 1}
});
Distribution of noise.
variable stopping_cond_defs¶
std::vector< std::string > stopping_cond_defs;
Stopping condition PolicyChain definitions.
variable defensibility¶
std::optional< float > defensibility;
The defensibility factor of the strategy.
variable prevalence¶
std::optional< float > prevalence;
The prevalence factor of the strategy.
variable stage¶
HackingStage stage {HackingStage::PostProcessing};
Updated on 29 June 2021 at 16:13:47 CEST
Last update: 2021-09-18