Fairness PCA

Calculate PC for metric_matrix to see similarities between models and metrics. If omit_models_with_NA is set to TRUE models with NA will be omitted as opposed to default behavior, when metrics are omitted.

fairness_pca(x, omit_models_with_NA = FALSE)

Arguments

x	object of class fairness object
omit_models_with_NA	logical, if TRUE omits rows in metric_matrix, else omits columns (default)

Value

fairness_pca object It is list containing following fields:

• pc_1_2 - amount of data variance explained with each component

• rotation - rotation from stats::prcomp

• x - x from stats::prcomp

• sdev - sdev from stats::prcomp

• label - model labels

Examples

data("german")

y_numeric <- as.numeric(german$Risk) - 1

lm_model <- glm(Risk ~ .,
  data = german,
  family = binomial(link = "logit")
)

rf_model <- ranger::ranger(Risk ~ .,
  data = german,
  probability = TRUE,
  num.trees = 200,
  num.threads = 1
)

explainer_lm <- DALEX::explain(lm_model, data = german[, -1], y = y_numeric)
#> Preparation of a new explainer is initiated
#>   -> model label       :  lm  (  default  )
#>   -> data              :  1000  rows  9  cols 
#>   -> target variable   :  1000  values 
#>   -> predict function  :  yhat.glm  will be used (  default  )
#>   -> predicted values  :  No value for predict function target column. (  default  )
#>   -> model_info        :  package stats , ver. 4.1.1 , task classification (  default  ) 
#>   -> predicted values  :  numerical, min =  0.1369187 , mean =  0.7 , max =  0.9832426  
#>   -> residual function :  difference between y and yhat (  default  )
#>   -> residuals         :  numerical, min =  -0.9572803 , mean =  1.940006e-17 , max =  0.8283475  
#>   A new explainer has been created!  
explainer_rf <- DALEX::explain(rf_model, data = german[, -1], y = y_numeric)
#> Preparation of a new explainer is initiated
#>   -> model label       :  ranger  (  default  )
#>   -> data              :  1000  rows  9  cols 
#>   -> target variable   :  1000  values 
#>   -> predict function  :  yhat.ranger  will be used (  default  )
#>   -> predicted values  :  No value for predict function target column. (  default  )
#>   -> model_info        :  package ranger , ver. 0.13.1 , task classification (  default  ) 
#>   -> predicted values  :  numerical, min =  0.05815476 , mean =  0.6969663 , max =  0.9975069  
#>   -> residual function :  difference between y and yhat (  default  )
#>   -> residuals         :  numerical, min =  -0.7217679 , mean =  0.003033675 , max =  0.6242611  
#>   A new explainer has been created!  

fobject <- fairness_check(explainer_lm, explainer_rf,
  protected = german$Sex,
  privileged = "male"
)
#> Creating fairness classification object
#> -> Privileged subgroup		: character ( Ok  )
#> -> Protected variable		: factor ( Ok  ) 
#> -> Cutoff values for explainers	: 0.5 ( for all subgroups ) 
#> -> Fairness objects		: 0 objects 
#> -> Checking explainers		: 2 in total (  compatible  )
#> -> Metric calculation		: 10/13 metrics calculated for all models ( 3 NA created )
#>  Fairness object created succesfully  

# same explainers with different cutoffs for female
fobject <- fairness_check(explainer_lm, explainer_rf, fobject,
  protected = german$Sex,
  privileged = "male",
  cutoff = list(female = 0.4),
  label = c("lm_2", "rf_2")
)
#> Creating fairness classification object
#> -> Privileged subgroup		: character ( Ok  )
#> -> Protected variable		: factor ( Ok  ) 
#> -> Cutoff values for explainers	: female: 0.4, male: 0.5 
#> -> Fairness objects		: 1 object (  compatible  )
#> -> Checking explainers		: 4 in total (  compatible  )
#> -> Metric calculation		: 10/13 metrics calculated for all models ( 3 NA created )
#>  Fairness object created succesfully  

fpca <- fairness_pca(fobject)
#> Warning: Found metric with NA: FNR, FOR, NEW_METRIC, omiting it

plot(fpca)