Metric

    Contains

    The metric values for the training and test sets.

    The table below lists the names of parameters that define the metric values to output. The values of all functions defined by these parameters are output.

    Command-line version parameters Python parameters R parameters
    --custom-metric custom_metric custom_loss
    --loss-function --loss-function --loss-function
    --eval-metric --eval-metric --eval-metric

    Format

    • The first row describes the data provided in the file.

      Format:

      iter<\t><loss 1><loss 2><\t>...<\t><loss N>
      
    • The metric names are expanded by colon-separated numbers if several validation datasets are input. The numbers correspond to the serial number of the input dataset.

    • All the rows except the first contain information for the specific iteration of building the tree.

      Format:

      <tree index><\t><loss 1><loss 2><\t>...<\t><loss N>
      

    Examples

    iter<\t>Logloss<\t>AUC
    0<\t>0.6637258841<\t>0.8800403474
    1<\t>0.6358649829<\t>0.8898645092
    2<\t>0.6118586328<\t>0.8905880184
    3<\t>0.5882755767<\t>0.8911104564
    4<\t>0.5665035887<\t>0.8933460724
    

    The output for three input validation datasets:

    iter<\t>RMSE<\t>RMSE:1<\t>RMSE:2
    0<\t>0.114824346<\t>0.1105841934<\t>0.08683344953
    1<\t>0.1136556268<\t>0.1095536596<\t>0.08584400666
    2<\t>0.1125784149<\t>0.10852689<\t>0.08494974738
    3<\t>0.1114784956<\t>0.1075251632<\t>0.08401943147
    4<\t>0.1103751142<\t>0.106557555<\t>0.08312388916