Parameter tuning

Sometimes when categorical features don't have a lot of values, one-hot encoding works well.

It is recommended to check that there is no obvious underfitting or overfitting before tuning any other parameters. In order to do this it is necessary to analyze the metric value on the validation dataset and select the appropriate number of iterations.

This can be done by setting the number of iterations to a large value, using the overfitting detector parameters and turning the use best model options on. In this case the resulting model contains only the first k best iterations, where k is the iteration with the best loss value on the validation dataset.

Also, the metric for choosing the best model may differ from the one used for optimizing the objective value. For example, it is possible to set the optimized function to Logloss and use the AUC function for the overfitting detector. To do so, use the evaluation metric parameter.

Parameters

Command-line version parameters	Python parameters	R parameters	Description
-i --iterations	iterations	iterations	The maximum number of trees that can be built when solving machine learning problems. When using other parameters that limit the number of iterations, the final number of trees may be less than the number specified in this parameter.
--use-best-model	use_best_model	use_best_model	If this parameter is set, the number of trees that are saved in the resulting model is defined as follows: Build the number of trees defined by the training parameters. Use the validation dataset to identify the iteration with the optimal value of the metric specified in  --eval-metric (eval_metric). No trees are saved after this iteration. This option requires a validation dataset to be provided.
--eval-metric	eval_metric	eval_metric	The metric used for overfitting detection (if enabled) and best model selection (if enabled). Some metrics support optional parameters (see the Objectives and metrics section for details on each metric). Format: <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] 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 NormalizedGini PrecisionAt RecallAt MAP Examples: R2 Quantile:alpha=0.3
Overfitting detection settings
--od-type	od_type	od_type	The type of the overfitting detector to use. Possible values: IncToDec Iter
--od-pval	od_pval	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. For best results, it is recommended to set a value in the range . The larger the value, the earlier overfitting is detected. Restriction. Do not use this parameter with the Iter overfitting detector type.
--od-wait	od_wait	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: IncToDec — Ignore the overfitting detector when the threshold is reached and continue learning for the specified number of iterations after the iteration with the optimal metric value. Iter — Consider the model overfitted and stop training after the specified number of iterations since the iteration with the optimal metric value.

Command-line version parameters	Python parameters	R parameters	Description
-i --iterations	iterations	iterations	The maximum number of trees that can be built when solving machine learning problems. When using other parameters that limit the number of iterations, the final number of trees may be less than the number specified in this parameter.
--use-best-model	use_best_model	use_best_model	If this parameter is set, the number of trees that are saved in the resulting model is defined as follows: Build the number of trees defined by the training parameters. Use the validation dataset to identify the iteration with the optimal value of the metric specified in  --eval-metric (eval_metric). No trees are saved after this iteration. This option requires a validation dataset to be provided.
--eval-metric	eval_metric	eval_metric	The metric used for overfitting detection (if enabled) and best model selection (if enabled). Some metrics support optional parameters (see the Objectives and metrics section for details on each metric). Format: <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] 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 NormalizedGini PrecisionAt RecallAt MAP Examples: R2 Quantile:alpha=0.3
Overfitting detection settings
--od-type	od_type	od_type	The type of the overfitting detector to use. Possible values: IncToDec Iter
--od-pval	od_pval	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. For best results, it is recommended to set a value in the range . The larger the value, the earlier overfitting is detected. Restriction. Do not use this parameter with the Iter overfitting detector type.
--od-wait	od_wait	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: IncToDec — Ignore the overfitting detector when the threshold is reached and continue learning for the specified number of iterations after the iteration with the optimal metric value. Iter — Consider the model overfitted and stop training after the specified number of iterations since the iteration with the optimal metric value.

This setting is used for reducing the gradient step. It affects the overall time of training: the smaller the value, the more iterations are required for training. Choose the value based on the performance expectations.

By default, the learning rate is defined automatically based on the dataset properties and the number of iterations. The automatically defined value should be close to the optimal one.

Parameters

Command-line version parameters	Python parameters	R parameters	Description
-w --learning-rate	learning_rate	learning_rate	The learning rate. Used for reducing the gradient step.

Command-line version parameters	Python parameters	R parameters	Description
-w --learning-rate	learning_rate	learning_rate	The learning rate. Used for reducing the gradient step.

In most cases, the optimal depth ranges from 4 to 10. Values in the range from 6 to 10 are recommended.

Parameters

Command-line version parameters	Python parameters	R parameters	Description
-n --depth	depth	depth	Depth of the tree. The range of supported values depends on the processing unit type and the type of the selected loss function: CPU — Any integer up to  16. GPU — Any integer up to 8 pairwise modes (YetiRank, PairLogitPairwise and QueryCrossEntropy) and up to   16 for all other loss functions.

Command-line version parameters	Python parameters	R parameters	Description
-n --depth	depth	depth	Depth of the tree. The range of supported values depends on the processing unit type and the type of the selected loss function: CPU — Any integer up to  16. GPU — Any integer up to 8 pairwise modes (YetiRank, PairLogitPairwise and QueryCrossEntropy) and up to   16 for all other loss functions.

Try different values for the regularizer to find the best possible.

Parameters

Command-line version parameters	Python parameters	R parameters	Description
--l2-leaf-reg	l2_leaf_reg	l2_leaf_reg	Coefficient at the L2 regularization term of the cost function. Any positive value is allowed.

Command-line version parameters	Python parameters	R parameters	Description
--l2-leaf-reg	l2_leaf_reg	l2_leaf_reg	Coefficient at the L2 regularization term of the cost function. Any positive value is allowed.

Try setting different values for the random_strength parameter.

Parameters

Command-line version parameters	Python parameters	R parameters	Description
--random-strength	random_strength	random_strength	The amount of randomness to use for scoring splits when the tree structure is selected. Use this parameter to avoid overfitting the model. The value of this parameter is used when selecting splits. On every iteration each possible split gets a score (for example, the score indicates how much adding this split will improve the loss function for the training dataset). The split with the highest score is selected. The scores have no randomness. A normally distributed random variable is added to the score of the feature. It has a zero mean and a variance that decreases during the training. The value of this parameter is the multiplier of the variance. Note. This parameter is not supported for the following loss functions: QueryCrossEntropy YetiRankPairwise PairLogitPairwise

Command-line version parameters	Python parameters	R parameters	Description
--random-strength	random_strength	random_strength	The amount of randomness to use for scoring splits when the tree structure is selected. Use this parameter to avoid overfitting the model. The value of this parameter is used when selecting splits. On every iteration each possible split gets a score (for example, the score indicates how much adding this split will improve the loss function for the training dataset). The split with the highest score is selected. The scores have no randomness. A normally distributed random variable is added to the score of the feature. It has a zero mean and a variance that decreases during the training. The value of this parameter is the multiplier of the variance. Note. This parameter is not supported for the following loss functions: QueryCrossEntropy YetiRankPairwise PairLogitPairwise

Try setting different values for the bagging_temperature parameter

Parameters

Command-line version parameters	Python parameters	R parameters	Description
--bagging-temperature	bagging_temperature	bagging_temperature	Defines the settings of the Bayesian bootstrap. It is used by default in classification and regression modes. Use the Bayesian bootstrap to assign random weights to objects. The weights are sampled from exponential distribution if the value of this parameter is set to “1”. All weights are equal to 1 if the value of this parameter is set to “0”. Possible values are in the range . The higher the value the more aggressive the bagging is. This parameter can be used if the selected bootstrap type is Bayesian.

Command-line version parameters	Python parameters	R parameters	Description
--bagging-temperature	bagging_temperature	bagging_temperature	Defines the settings of the Bayesian bootstrap. It is used by default in classification and regression modes. Use the Bayesian bootstrap to assign random weights to objects. The weights are sampled from exponential distribution if the value of this parameter is set to “1”. All weights are equal to 1 if the value of this parameter is set to “0”. Possible values are in the range . The higher the value the more aggressive the bagging is. This parameter can be used if the selected bootstrap type is Bayesian.

The number of splits for numerical features.

By default, it is set to 254 (if training is performed on CPU) or 128 (if training is performed on GPU).

The value of this parameter significantly impacts the speed of training on GPU. The smaller the value, the faster the training is performed (refer to the Number of splits for numerical features section for details).

128 splits are enough for many datasets. However, try to set the value of this parameter to 254 when training on GPU if the best possible quality is required.

The value of this parameter does not significantly impact the speed of training on CPU. Try to set it to 254 for the best possible quality.

Parameters

Command-line version parameters	Python parameters	R parameters	Description
-x --border-count	border_count Alias: max_bin	border_count	The number of splits for numerical features. Allowed values depend on the processing unit type: CPU — integers from 1 to 65535 inclusively. GPU — integers from 1 to 255 inclusively. Recommended values are up to 255. Larger values slow down the training.

Command-line version parameters	Python parameters	R parameters	Description
-x --border-count	border_count Alias: max_bin	border_count	The number of splits for numerical features. Allowed values depend on the processing unit type: CPU — integers from 1 to 65535 inclusively. GPU — integers from 1 to 255 inclusively. Recommended values are up to 255. Larger values slow down the training.

Use this option if the objects in your dataset are given in the required order. In this case, random permutations are not performed during the Transforming categorical features to numerical features and Choosing the tree structure stages.

Parameters

Command-line version parameters	Python parameters	R parameters	Description
--has-time	has_time	has_time	Use the order of objects in the input data (do not perform random permutations during the Transforming categorical features to numerical features and Choosing the tree structure stages). The Timestamp column type is used to determine the order of objects if specified in the input data.

Command-line version parameters	Python parameters	R parameters	Description
--has-time	has_time	has_time	Use the order of objects in the input data (do not perform random permutations during the Transforming categorical features to numerical features and Choosing the tree structure stages). The Timestamp column type is used to determine the order of objects if specified in the input data.

By default, CatBoost uses symmetric trees, which are built if the growing policy is set to SymmetricTree.

Such trees are built level by level until the specified depth is reached. On each iteration, all leaves from the last tree level are split with the same condition. The resulting tree structure is always symmetric.

Symmetric trees have a very good prediction speed (roughly 10 times faster than regular trees) and give better quality in many cases.

However, in some cases, other tree growing strategies can give better results than growing symmetric trees.

Try to analyze the results obtained with different growing trees strategies.

Parameters

Command-line version parameters	Python parameters	R parameters	Description
--grow-policy	grow_policy	grow_policy	The tree growing policy. Defines how to perform greedy tree construction. Possible values: SymmetricTree —A tree is built level by level until the specified depth is reached. On each iteration, all leaves from the last tree level are split with the same condition. The resulting tree structure is always symmetric. Depthwise — A tree is built level by level until the specified depth is reached. On each iteration, all non-terminal leaves from the last tree level are split. Each leaf is split by condition with the best loss improvement. Lossguide — A tree is built leaf by leaf until the specified maximum number of leaves is reached. On each iteration, non-terminal leaf with the best loss improvement is split. Note. The Depthwise and Lossguide growing policies are currently supported only in training and prediction modes. They are not supported for model analysis (such as Feature importance and ShapValues) and exporting to different model formats (such as AppleCoreML , onnx and json) .
--min-data-in-leaf	min_data_in_leaf Alias: min_child_samples	min_data_in_leaf	The minimum number of training samples in a leaf. CatBoost does not search for new splits in leaves with samples count less than the specified value. Can be used only with the Lossguide and Depthwise growing policies.
--max-leaves	max_leaves Alias: num_leaves	max_leaves	The maximum number of leafs in the resulting tree. Can be used only with the Lossguide growing policy. Tip. It is not recommended to use values greater than 64, since it can significantly slow down the training process.

Command-line version parameters	Python parameters	R parameters	Description
--grow-policy	grow_policy	grow_policy	The tree growing policy. Defines how to perform greedy tree construction. Possible values: SymmetricTree —A tree is built level by level until the specified depth is reached. On each iteration, all leaves from the last tree level are split with the same condition. The resulting tree structure is always symmetric. Depthwise — A tree is built level by level until the specified depth is reached. On each iteration, all non-terminal leaves from the last tree level are split. Each leaf is split by condition with the best loss improvement. Lossguide — A tree is built leaf by leaf until the specified maximum number of leaves is reached. On each iteration, non-terminal leaf with the best loss improvement is split. Note. The Depthwise and Lossguide growing policies are currently supported only in training and prediction modes. They are not supported for model analysis (such as Feature importance and ShapValues) and exporting to different model formats (such as AppleCoreML , onnx and json) .
--min-data-in-leaf	min_data_in_leaf Alias: min_child_samples	min_data_in_leaf	The minimum number of training samples in a leaf. CatBoost does not search for new splits in leaves with samples count less than the specified value. Can be used only with the Lossguide and Depthwise growing policies.
--max-leaves	max_leaves Alias: num_leaves	max_leaves	The maximum number of leafs in the resulting tree. Can be used only with the Lossguide growing policy. Tip. It is not recommended to use values greater than 64, since it can significantly slow down the training process.