eval_metrics

Calculate the specified metrics for the specified dataset.

Method call format

eval_metrics(data,
             metrics,
             ntree_start=0,
             ntree_end=0,
             eval_period=1,
             thread_count=-1)

Parameters

Parameter Possible values Description Default value
data catboost.Pool

A file or matrix with the input dataset.

Required parameter
metrics list of strings

The list of metrics to be calculated.

Supported metrics
  • RMSE
  • Logloss
  • MAE
  • CrossEntropy
  • Quantile
  • LogLinQuantile
  • Lq
  • MultiClass
  • MultiClassOneVsAll
  • MAPE
  • Poisson
  • PairLogit
  • PairLogitPairwise
  • QueryRMSE
  • QuerySoftMax
  • SMAPE
  • Recall
  • Precision
  • F1
  • TotalF1
  • Accuracy
  • BalancedAccuracy
  • BalancedErrorRate
  • Kappa
  • WKappa
  • LogLikelihoodOfPrediction
  • AUC
  • R2
  • FairLoss
  • NumErrors
  • MCC
  • BrierScore
  • HingeLoss
  • HammingLoss
  • ZeroOneLoss
  • MSLE
  • MedianAbsoluteError
  • Huber
  • Expectile
  • PairAccuracy
  • AverageGain
  • PFound
  • NDCG
  • DCG
  • FilteredDCG<