Using the overfitting detector
If overfitting occurs, CatBoost can stop the training earlier than the training parameters dictate. For example, it can be stopped before the specified number of trees are built. This option is set in the starting parameters.
Choose the implementation for more details.
The following parameters can be set for the corresponding methods and are used when the model is trained.
| Method | Parameters | Description |
|---|---|---|
| od_type | The type of the overfitting detector to use. Possible values:
| |
| od_pval | The threshold for the IncToDec overfitting detector type. The training is stopped when the specified value is reached. Requires that a validation dataset was input. The value of this parameter must exceed “0” to turn on the overfitting detection. | |
| od_wait | The number of iterations to continue the training after the iteration with the optimal metric value. The purpose of this parameter differs depending on the selected overfitting detector type:
|
| Method | Parameters | Description |
|---|---|---|
| od_type | The type of the overfitting detector to use. Possible values:
| |
| od_pval | The threshold for the IncToDec overfitting detector type. The training is stopped when the specified value is reached. Requires that a validation dataset was input. The value of this parameter must exceed “0” to turn on the overfitting detection. | |
| od_wait | The number of iterations to continue the training after the iteration with the optimal metric value. The purpose of this parameter differs depending on the selected overfitting detector type:
|
The following parameters can be set for the corresponding methods and are used when the model is trained.
| Method | Parameters | Description |
|---|---|---|
| catboost.train | od_type | The type of the overfitting detector to use. Possible values:
|
| od_pval | The threshold for the IncToDec overfitting detector type. The training is stopped when the specified value is reached. Requires that a validation dataset was input. The value of this parameter must exceed “0” to turn on the overfitting detection. | |
| od_wait | The number of iterations to continue the training after the iteration with the optimal metric value. The purpose of this parameter differs depending on the selected overfitting detector type:
|
| Method | Parameters | Description |
|---|---|---|
| catboost.train | od_type | The type of the overfitting detector to use. Possible values:
|
| od_pval | The threshold for the IncToDec overfitting detector type. The training is stopped when the specified value is reached. Requires that a validation dataset was input. The value of this parameter must exceed “0” to turn on the overfitting detection. | |
| od_wait | The number of iterations to continue the training after the iteration with the optimal metric value. The purpose of this parameter differs depending on the selected overfitting detector type:
|
| Command | Command keys | Key description |
|---|---|---|
| catboost fit | --od-type | The type of the overfitting detector to use. Possible values:
|
| --od-pval | The threshold for the IncToDec overfitting detector type. The training is stopped when the specified value is reached. Requires that a validation dataset was input. The value of this parameter must exceed “0” to turn on the overfitting detection. | |
| --od-wait | The number of iterations to continue the training after the iteration with the optimal metric value. The purpose of this parameter differs depending on the selected overfitting detector type:
|
| Command | Command keys | Key description |
|---|---|---|
| catboost fit | --od-type | The type of the overfitting detector to use. Possible values:
|
| --od-pval | The threshold for the IncToDec overfitting detector type. The training is stopped when the specified value is reached. Requires that a validation dataset was input. The value of this parameter must exceed “0” to turn on the overfitting detection. | |
| --od-wait | The number of iterations to continue the training after the iteration with the optimal metric value. The purpose of this parameter differs depending on the selected overfitting detector type:
|