perf_eva: Binomial Metrics

Description

perf_eva calculates metrics to evaluate the performance of binomial classification model. It can also creates confusion matrix and model performance graphics.

Usage

perf_eva(pred, label, title = NULL, binomial_metric = c("mse", "rmse",
  "logloss", "r2", "ks", "auc", "gini"), confusion_matrix = TRUE,
  threshold = NULL, show_plot = c("ks", "roc"), positive = "bad|1",
  ...)

Arguments

pred

A list or vector of predicted probability or score.

label

A list or vector of label values.

title

The title of plot. Default is NULL.

binomial_metric

Default is c('mse', 'rmse', 'logloss', 'r2', 'ks', 'auc', 'gini'). If it is NULL, then no metric will calculated.

confusion_matrix

Logical, whether to create a confusion matrix. Default is TRUE.

threshold

Confusion matrix threshold. Default is the pred on maximum F1.

show_plot

Default is c('ks', 'roc'). Accepted values including c('ks', 'lift', 'gain', 'roc', 'lz', 'pr', 'f1', 'density').

positive

Value of positive class, default is "bad|1".

...

Additional parameters.

Value

A list of binomial metric, confusion matrix and graphics

Details

Accuracy = true positive and true negative/total cases

Error rate = false positive and false negative/total cases

TPR, True Positive Rate(Recall or Sensitivity) = true positive/total actual positive

PPV, Positive Predicted Value(Precision) = true positive/total predicted positive

TNR, True Negative Rate(Specificity) = true negative/total actual negative = 1-FPR

NPV, Negative Predicted Value = true negative/total predicted negative

Examples

Run this code

# NOT RUN {
# data preparing ------
# load germancredit data
data("germancredit")
# filter variable via missing rate, iv, identical value rate
dt_f = var_filter(germancredit, "creditability")
# breaking dt into train and test
dt_list = split_df(dt_f, "creditability")
label_list = lapply(dt_list, function(x) x$creditability)

# woe binning ------
bins = woebin(dt_list$train, "creditability")
# converting train and test into woe values
dt_woe_list = lapply(dt_list, function(x) woebin_ply(x, bins))

# glm ------
m1 = glm(creditability ~ ., family = binomial(), data = dt_woe_list$train)
# vif(m1, merge_coef = TRUE)
# Select a formula-based model by AIC
m_step = step(m1, direction="both", trace=FALSE)
m2 = eval(m_step$call)
# vif(m2, merge_coef = TRUE)

# predicted proability
pred_list = lapply(dt_woe_list, function(x) predict(m2, type = 'response', x))

# scorecard ------
card = scorecard(bins, m2)

# credit score, only_total_score = TRUE
score_list = lapply(dt_list, function(x) scorecard_ply(x, card))
# credit score, only_total_score = FALSE
score_list2 = lapply(dt_list, function(x) scorecard_ply(x, card, only_total_score=FALSE))


###### perf_eva examples ######
# Example I, one datset
## predicted p1
perf_eva(pred = pred_list$train, label=dt_list$train$creditability, title = 'train')
## predicted score
# perf_eva(pred = score_list$train, label=dt_list$train$creditability, title = 'train')

# Example II, multiple datsets
## predicted p1
perf_eva(pred = pred_list, label = label_list)
## predicted score
# perf_eva(score_list, label_list)


###### perf_psi examples ######
# Example I # only total psi
psi1 = perf_psi(score = score_list, label = label_list)
psi1$psi  # psi data frame
psi1$pic  # pic of score distribution

# Example II # both total and variable psi
psi2 = perf_psi(score = score_list, label = label_list)
# psi2$psi  # psi data frame
# psi2$pic  # pic of score distribution


###### gains_table examples ######
# Example I, input score and label can be a list or a vector
gains_table(score = score_list$train, label = label_list$train)
gains_table(score = score_list, label = label_list)

# Example II, specify the bins number and type
gains_table(score = score_list, label = label_list, bin_num = 20)
gains_table(score = score_list, label = label_list, bin_type = 'width')
# }
# NOT RUN {
# }