get_roc_curve

Return points of the ROC curve.

This information is used to plot the ROC curve.

Method call format

get_roc_curve(model,
              data,
              thread_count=-1,
              plot=False)

Parameters

model

Description

The trained model.

Possible types

catboost.CatBoost

Default value

Required parameter

data

Description

A set of samples to build the ROC curve with.

Possible types

  • catboost.Pool
  • list of catboost.Pool

Default value

Required parameter

thread_count

Description

The number of threads to use.

Optimizes the speed of execution. This parameter doesn't affect results.

Possible type

int

Default value

-1 (the number of threads is equal to the number of processor cores)

plot

Description

Plot a chart based on the found points.

Possible types

bool

Default value

False

Type of return value

tuple of three arrays (fpr, tpr, thresholds)

Usage examples

from catboost import CatBoostClassifier, Pool
from catboost.utils import get_roc_curve

train_data = [[1,3],
              [0,4],
              [1,7],
              [0,3]]
train_labels = [1,0,1,1]
catboost_pool = Pool(train_data, train_labels)
model = CatBoostClassifier(learning_rate=0.03)
model.fit(train_data, train_labels, verbose=False)
(fpr, tpr, thresholds) = get_roc_curve(model, catboost_pool, plot=True)
print(fpr)
print(tpr)
print(thresholds)

Output:

[0. 0. 0. 0. 1.]
[0.         0.33333333 0.66666667 1.         1.        ]
[1.         0.53533186 0.52910032 0.50608183 0.        ]