---
metadata:
  - name: generator
    content: Diplodoc Platform v5.50.3
alternate:
  - https://catboost.ai/docs/en/references/training-parameters/index.md
---
> **Documentation Index:** Fetch the complete configuration index at https://catboost.ai/docs/en/llms.txt

# Overview

These parameters are for the Python package, R package and Command-line version.

<!-- source: en/_includes/work_src/reusage-python/how-aliases-are-applied-intro.md -->
For the Python package several parameters have aliases. For example, the `--iterations` parameter has the following synonyms: `num_boost_round`, `n_estimators`, `num_trees`. Simultaneous usage of different names of one parameter raises an error.
<!-- endsource: en/_includes/work_src/reusage-python/how-aliases-are-applied-intro.md -->

<!-- source: en/_includes/work_src/reusage-code-examples/nvidia-driver-reqs.md -->
Training or inference on CUDA-enabled GPUs requires NVIDIA Driver of version 450.80.02 or higher.
<!-- endsource: en/_includes/work_src/reusage-code-examples/nvidia-driver-reqs.md -->

## Common parameters

### [loss_function](https://catboost.ai/docs/en/references/training-parameters/common.md#loss_function)

Command-line: `--loss-function`

_Alias:_ `objective`

<!-- source: en/_includes/work_src/reusage/loss-function-short-desc.md -->
The [metric](https://catboost.ai/docs/en/concepts/loss-functions.md) to use in training. The specified value also determines the machine learning problem to solve. Some metrics support optional parameters (see the [Objectives and metrics](https://catboost.ai/docs/en/concepts/loss-functions.md) section for details on each metric).
<!-- endsource: en/_includes/work_src/reusage/loss-function-short-desc.md -->

### [custom_metric](https://catboost.ai/docs/en/references/training-parameters/common.md#custom_metric)

Command-line: `--custom-metric`

<!-- source: en/_includes/work_src/reusage/custom-loss--basic.md -->
[Metric](https://catboost.ai/docs/en/concepts/loss-functions.md) values to output during training. These functions are not optimized and are displayed for informational purposes only. Some metrics support optional parameters (see the [Objectives and metrics](https://catboost.ai/docs/en/concepts/loss-functions.md) section for details on each metric).
<!-- endsource: en/_includes/work_src/reusage/custom-loss--basic.md -->

### [eval_metric](https://catboost.ai/docs/en/references/training-parameters/common.md#eval_metric)

Command-line: `--eval-metric`

<!-- source: en/_includes/work_src/reusage/eval-metric--basic.md -->
The metric used for overfitting detection (if enabled) and best model selection (if enabled). Some metrics support optional parameters (see the [Objectives and metrics](https://catboost.ai/docs/en/concepts/loss-functions.md) section for details on each metric).
<!-- endsource: en/_includes/work_src/reusage/eval-metric--basic.md -->

### [iterations](https://catboost.ai/docs/en/references/training-parameters/common.md#iterations)

Command-line: `-i`, `--iterations`

_Aliases:_ `num_boost_round`, `n_estimators`, `num_trees`

The maximum number of trees that can be built when solving machine learning problems.

### [learning_rate](https://catboost.ai/docs/en/references/training-parameters/common.md#learning_rate)

Command-line: `-w`, `--learning-rate`

_Alias:_ `eta`

The learning rate.

Used for reducing the gradient step.

### [random_seed](https://catboost.ai/docs/en/references/training-parameters/common.md#random_seed)

Command-line: `-r`, `--random-seed`

_Alias:_`random_state`

The random seed used for training.

### [l2_leaf_reg](https://catboost.ai/docs/en/references/training-parameters/common.md#l2_leaf_reg)

Command-line: `--l2-leaf-reg`, `l2-leaf-regularizer`

_Alias:_ `reg_lambda`

Coefficient at the L2 regularization term of the cost function.

### [bootstrap_type](https://catboost.ai/docs/en/references/training-parameters/common.md#bootstrap_type)

Command-line: `--bootstrap-type`

[Bootstrap type](https://catboost.ai/docs/en/concepts/algorithm-main-stages_bootstrap-options.md). Defines the method for sampling the weights of objects.

### [bagging_temperature](https://catboost.ai/docs/en/references/training-parameters/common.md#bagging_temperature)

Command-line: `--bagging-temperature`

Defines the settings of the Bayesian bootstrap. It is used by default in classification and regression modes.

### [subsample](https://catboost.ai/docs/en/references/training-parameters/common.md#subsample)

Command-line: `--subsample`

Sample rate for bagging.

### [sampling_frequency](https://catboost.ai/docs/en/references/training-parameters/common.md#sampling_frequency)

Command-line: `--sampling-frequency`

Frequency to sample weights and objects when building trees.

### [sampling_unit](https://catboost.ai/docs/en/references/training-parameters/common.md#sampling_unit)

Command-line: `--sampling-unit`

The sampling scheme.

### [mvs_reg](https://catboost.ai/docs/en/references/training-parameters/common.md#mvs_reg)

Command-line: `--mvs-reg`

<!-- source: en/_includes/work_src/reusage/cli__mvs-head-fraction__div.md -->
Affects the weight of the denominator and can be used for balancing between the importance and Bernoulli sampling (setting it to 0 implies importance sampling and to $\infty$ - Bernoulli).
<!-- endsource: en/_includes/work_src/reusage/cli__mvs-head-fraction__div.md -->

### [random_strength](https://catboost.ai/docs/en/references/training-parameters/common.md#random_strength)

Command-line: `--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.

### [use_best_model](https://catboost.ai/docs/en/references/training-parameters/common.md#use_best_model)

Command-line: `--use-best-model`

Use the validation dataset to identify the iteration with the optimal value of the evaluation metric.
Save only trees up to this iteration (inclusive) in the resulting model.

### [best_model_min_trees](https://catboost.ai/docs/en/references/training-parameters/common.md#best_model_min_trees)

Command-line: `--best-model-min-trees`

<!-- source: en/_includes/work_src/reusage/clii__best-model-min-trees__short-desc.md -->
The minimal number of trees that the best model should have. If set, the output model contains at least the given number of trees even if the optimal value of the evaluation metric on the validation dataset is achieved with smaller number of trees.
<!-- endsource: en/_includes/work_src/reusage/clii__best-model-min-trees__short-desc.md -->

### [depth](https://catboost.ai/docs/en/references/training-parameters/common.md#depth)

Command-line: `-n`, `--depth`

_Alias:_ `max_depth`

Depth of the trees.

### [grow_policy](https://catboost.ai/docs/en/references/training-parameters/common.md#grow_policy)

Command-line: `--grow-policy`

The tree growing policy. Defines how to perform greedy tree construction.

### [min_data_in_leaf](https://catboost.ai/docs/en/references/training-parameters/common.md#min_data_in_leaf)

Command-line: `--min-data-in-leaf`

_Alias:_ `min_child_samples`

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.

### [max_leaves](https://catboost.ai/docs/en/references/training-parameters/common.md#max_leavescommon.md#)

Command-line: `--max-leaves`

_Alias:_`num_leaves`

The maximum number of leafs in the resulting tree. Can be used only with the Lossguide growing policy.

### [ignored_features](https://catboost.ai/docs/en/references/training-parameters/common.md#ignored_features)

Command-line: `-I`, `--ignore-features`

<!-- source: en/_includes/work_src/reusage/ignored-feature__common-div.md -->
Feature indices or names to exclude from the training. It is assumed that all passed values are feature names if at least one of the passed values can not be converted to a number or a range of numbers. Otherwise, it is assumed that all passed values are feature indices.

Specifics:

- Non-negative indices that do not match any features are successfully ignored. For example, if five features are defined for the objects in the dataset and this parameter is set to <q>42</q>, the corresponding non-existing feature is successfully ignored.

- The identifier corresponds to the feature's index. Feature indices used in train and feature importance are numbered from 0 to `featureCount – 1`. If a file is used as [input data](https://catboost.ai/docs/en/concepts/input-data.md) then any non-feature column types are ignored when calculating these indices. For example, each row in the input file contains data in the following order: `cat feature<\t>label value<\t>num feature`. So for the row `rock<\t>0<\t>42`, the identifier for the <q>rock</q> feature is 0, and for the <q>42</q> feature it's 1.


- The addition of a non-existing feature name raises an error.
<!-- endsource: en/_includes/work_src/reusage/ignored-feature__common-div.md -->

### [one_hot_max_size](https://catboost.ai/docs/en/references/training-parameters/common.md#one_hot_max_size)

Command-line: `--one-hot-max-size`

Use one-hot encoding for all categorical features with a number of different values less than or equal to the given parameter value. Ctrs are not calculated for such features.

### [has_time](https://catboost.ai/docs/en/references/training-parameters/common.md#has_time)

Command-line: `--has-time`

Use the order of objects in the input data (do not perform random permutations during the [Transforming categorical features to numerical features](https://catboost.ai/docs/en/concepts/algorithm-main-stages_cat-to-numberic.md) and [Choosing the tree structure](https://catboost.ai/docs/en/concepts/algorithm-main-stages_choose-tree-structure.md) stages).

### [rsm](https://catboost.ai/docs/en/references/training-parameters/common.md#rsm)

Command-line: `--rsm`

_Alias:_`colsample_bylevel`

Random subspace method. The percentage of features to use at each split selection, when features are selected over again at random.

### [nan_mode](https://catboost.ai/docs/en/references/training-parameters/common.md#nan_mode)

Command-line: `--nan-mode`

The method for  [processing missing values](https://catboost.ai/docs/en/concepts/algorithm-missing-values-processing.md) in the input dataset.

### [input_borders](https://catboost.ai/docs/en/references/training-parameters/common.md#input_borders)

Command-line: `--input-borders-file`

Load [Custom quantization borders and missing value modes](https://catboost.ai/docs/en/concepts/input-data_custom-borders.md) from a file (do not generate them).

### [output_borders](https://catboost.ai/docs/en/references/training-parameters/common.md#output_borders)

Command-line: `--output-borders-file`

Save quantization borders for the current dataset to a file.

### [fold_permutation_block](https://catboost.ai/docs/en/references/training-parameters/common.md#fold_permutation_block)

Command-line: `--fold-permutation-block`

Objects in the dataset are grouped in blocks before the random permutations. This parameter defines the size of the blocks.

### [leaf_estimation_method](https://catboost.ai/docs/en/references/training-parameters/common.md#leaf_estimation_method)

Command-line: `--leaf-estimation-method`

The method used to calculate the values in leaves.

### [leaf_estimation_iterations](https://catboost.ai/docs/en/references/training-parameters/common.md#leaf_estimation_iterations)

Command-line: `--leaf-estimation-iterations`

This parameter regulates how many steps are done in every tree when calculating leaf values.

### [leaf_estimation_backtracking](https://catboost.ai/docs/en/references/training-parameters/common.md#leaf_estimation_backtracking)

Command-line: `--leaf-estimation-backtracking`

When the value of the `leaf_estimation_iterations` parameter is greater than 1, CatBoost makes several gradient or newton steps when calculating the resulting leaf values of a tree.

### [fold_len_multiplier](https://catboost.ai/docs/en/references/training-parameters/common.md#fold_len_multiplier)

Command-line: `--fold-len-multiplier`

Coefficient for changing the length of folds.

### [approx_on_full_history](https://catboost.ai/docs/en/references/training-parameters/common.md#approx_on_full_history)

Command-line:`--approx-on-full-history`

The principles for calculating the approximated values.

### [class_weights](https://catboost.ai/docs/en/references/training-parameters/common.md#class_weights)

Command-line: `--class-weights`

<!-- source: en/_includes/work_src/reusage/class-weights__short-desc-intro.md -->
Class weights. The values are used as multipliers for the object weights. This parameter can be used for solving binary classification and multiclassification problems.
<!-- endsource: en/_includes/work_src/reusage/class-weights__short-desc-intro.md -->

### [class_names](https://catboost.ai/docs/en/references/training-parameters/common.md#class_names)

Classes names. Allows to redefine the default values when using the MultiClass and Logloss metrics.

### [auto_class_weights](https://catboost.ai/docs/en/references/training-parameters/common.md#auto_class_weights)

Command-line: `--auto-class-weights`

<!-- source: en/_includes/work_src/reusage/cli__auto-class-weights__div.md -->
Automatically calculate class weights based either on the total weight or the total number of objects in each class. The values are used as multipliers for the object weights.

Supported values:

- None — All class weights are set to 1
- Balanced:

    $CW_k=\displaystyle\frac{max_{c=1}^K(\sum_{t_{i}=c}{w_i})}{\sum_{t_{i}=k}{w_{i}}}$

- SqrtBalanced:

    $CW_k=\sqrt{\displaystyle\frac{max_{c=1}^K(\sum_{t_i=c}{w_i})}{\sum_{t_i=k}{w_i}}}$
<!-- endsource: en/_includes/work_src/reusage/cli__auto-class-weights__div.md -->

### [scale_pos_weight](https://catboost.ai/docs/en/references/training-parameters/common.md#scale_pos_weight)

The weight for class 1 in binary classification. The value is used as a multiplier for the weights of objects from class 1.

### [boosting_type](https://catboost.ai/docs/en/references/training-parameters/common.md#boosting_type)

Command-line: `--boosting-type`

Boosting scheme.

### [boost_from_average](https://catboost.ai/docs/en/references/training-parameters/common.md#boost_from_average)

Command-line: `--boost-from-average`

Initialize approximate values by best constant value for the specified loss function.

### [langevin](https://catboost.ai/docs/en/references/training-parameters/common.md#langevin)

Command-line: `--langevin`

Enables the Stochastic Gradient Langevin Boosting mode.

### [diffusion_temperature](https://catboost.ai/docs/en/references/training-parameters/common.md#diffusion_temperature)

Command-line: `--diffusion-temperature`

The diffusion temperature of the Stochastic Gradient Langevin Boosting mode.

### [posterior_sampling](https://catboost.ai/docs/en/references/training-parameters/common.md#posterior_sampling)

Command-line: `--posterior-sampling	`

If this parameter is set several options are specified as follows and model parameters are checked to obtain uncertainty predictions with good theoretical properties.

### [allow_const_label](https://catboost.ai/docs/en/references/training-parameters/common.md#allow_const_label)

Command-line: `--allow-const-label`

Use it to train models with datasets that have equal label values for all objects.

### [score_function](https://catboost.ai/docs/en/references/training-parameters/common.md#score_function)

Command-line: `--score-function`

The [score type](https://catboost.ai/docs/en/concepts/algorithm-score-functions.md) used to select the next split during the tree construction.

### [monotone_constraints](https://catboost.ai/docs/en/references/training-parameters/common.md#monotone_constraints)

Command-line: `--monotone-constraints`

<!-- source: en/_includes/work_src/reusage/cli__monotone-constraints__desc__div.md -->
Impose monotonic constraints on numerical features.

Possible values:

- <q>1</q> — Increasing constraint on the feature. The algorithm forces the model to be a non-decreasing function of this features.

- <q>-1</q> — Decreasing constraint on the feature. The algorithm forces the model to be a non-increasing function of this features.

- <q>0</q> — constraints are disabled.
<!-- endsource: en/_includes/work_src/reusage/cli__monotone-constraints__desc__div.md -->

### [feature_weights](https://catboost.ai/docs/en/references/training-parameters/common.md#feature_weights)

Command-line: `--feature-weights`

<!-- source: en/_includes/work_src/reusage/cli__feature-weight__desc__intro.md -->
Per-feature multiplication weights used when choosing the best split. The score of each candidate is multiplied by the weights of features from the current split.

Non-negative float values are supported for each weight.

Supported formats for setting the value of this parameter:
<!-- endsource: en/_includes/work_src/reusage/cli__feature-weight__desc__intro.md -->

### [first_feature_use_penalties](https://catboost.ai/docs/en/references/training-parameters/common.md#first_feature_use_penalties)

Command-line: `--first-feature-use-penalties`

<!-- source: en/_includes/work_src/reusage/cli__first-feature-use-penalties__intro.md -->
Per-feature penalties for the first occurrence of the feature in the model. The given value is subtracted from the score if the current candidate is the first one to include the feature in the model.

Refer to the [Per-object and per-feature penalties](https://catboost.ai/docs/en/concepts/algorithm-score-functions.md) section for details on applying different score penalties.

Non-negative float values are supported for each penalty.
<!-- endsource: en/_includes/work_src/reusage/cli__first-feature-use-penalties__intro.md -->

### [fixed_binary_splits](https://catboost.ai/docs/en/references/training-parameters/common.md#fixed_binary_splits)

Command-line: `--fixed-binary-splits`

A list of indices of binary features to put at the top of each tree.

### [penalties_coefficient](https://catboost.ai/docs/en/references/training-parameters/common.md#penalties_coefficient)

Command-line: `--penalties-coefficient`

A single-value common coefficient to multiply all penalties.

### [per_object_feature_penalties](https://catboost.ai/docs/en/references/training-parameters/common.md#per_object_feature_penalties)

Command-line: `--per-object-feature-penalties`

<!-- source: en/_includes/work_src/reusage/per-object-feature-penalties__intro.md -->
Per-object penalties for the first use of the feature for the object. The given value is multiplied by the number of objects that are divided by the current split and use the feature for the first time.

Refer to the [Per-object and per-feature penalties](https://catboost.ai/docs/en/concepts/algorithm-score-functions.md) section for details on applying different score penalties.

Non-negative float values are supported for each penalty.
<!-- endsource: en/_includes/work_src/reusage/per-object-feature-penalties__intro.md -->

### [model_shrink_rate](https://catboost.ai/docs/en/references/training-parameters/common.md#model_shrink_rate)

Command-line: `--model-shrink-rate`

The constant used to calculate the coefficient for multiplying the model on each iteration.

### [model_shrink_mode](https://catboost.ai/docs/en/references/training-parameters/common.md#model_shrink_mode)

Command-line: `model_shrink_mode`

Determines how the actual model shrinkage coefficient is calculated at each iteration.

## CTR settings

### [simple_ctr](https://catboost.ai/docs/en/references/training-parameters/ctr.md#simple_ctr)

<!-- source: en/_includes/work_src/reusage/cli__simple-ctr__intro.md -->
Quantization settings for simple [categorical features](https://catboost.ai/docs/en/concepts/algorithm-main-stages_cat-to-numberic.md). Use this parameter to specify the principles for defining the class of the object for regression tasks. By default, it is considered that an object belongs to the positive class if its' label value is greater than the median of all label values of the dataset.
<!-- endsource: en/_includes/work_src/reusage/cli__simple-ctr__intro.md -->

### [combinations_ctr](https://catboost.ai/docs/en/references/training-parameters/ctr.md#combinations_ctr)

<!-- source: en/_includes/work_src/reusage/cli__combination-ctr__intro.md -->
Quantization settings for combinations of [categorical features](https://catboost.ai/docs/en/concepts/algorithm-main-stages_cat-to-numberic.md).
<!-- endsource: en/_includes/work_src/reusage/cli__combination-ctr__intro.md -->

### [per_feature_ctr](https://catboost.ai/docs/en/references/training-parameters/ctr.md#per_feature_ctr)

<!-- source: en/_includes/work_src/reusage/cli__per-feature-ctr__intro.md -->
Per-feature quantization settings for [categorical features](https://catboost.ai/docs/en/concepts/algorithm-main-stages_cat-to-numberic.md).
<!-- endsource: en/_includes/work_src/reusage/cli__per-feature-ctr__intro.md -->

### [ctr_target_border_count](https://catboost.ai/docs/en/references/training-parameters/ctr.md#ctr_target_border_count)

<!-- source: en/_includes/work_src/reusage/cli__ctr-target-border-count__short-desc.md -->
The maximum number of borders to use in target quantization for categorical features that need it. Allowed values are integers from 1 to 255 inclusively.
<!-- endsource: en/_includes/work_src/reusage/cli__ctr-target-border-count__short-desc.md -->

### [counter_calc_method](https://catboost.ai/docs/en/references/training-parameters/ctr.md#counter_calc_method)

The method for calculating the Counter CTR type.

### [max_ctr_complexity](https://catboost.ai/docs/en/references/training-parameters/ctr.md#max_ctr_complexity)

The maximum number of features that can be combined.

### [ctr_leaf_count_limit](https://catboost.ai/docs/en/references/training-parameters/ctr.md#ctr_leaf_count_limit)

The maximum number of leaves with categorical features. If the quantity exceeds the specified value a part of leaves is discarded.

### [store_all_simple_ctr](https://catboost.ai/docs/en/references/training-parameters/ctr.md#store_all_simple_ctr)

Ignore categorical features, which are not used in feature combinations, when choosing candidates for exclusion.

### [final_ctr_computation_mode](https://catboost.ai/docs/en/references/training-parameters/ctr.md#final_ctr_computation_mode)

Final CTR computation mode.

## Input file settings

### [-f, --learn-set](https://catboost.ai/docs/en/references/training-parameters/input.md#-f)

The path to the input file  that contains the dataset.

### [-t, --test-set](https://catboost.ai/docs/en/references/training-parameters/input.md#-t)

A comma-separated list of input files that contain the validation dataset (the format must be the same as used in the training dataset).

### [--cd, --column-description](https://catboost.ai/docs/en/references/training-parameters/input.md#--cd)

The path to the input file  that contains the [columns description](https://catboost.ai/docs/en/concepts/input-data_column-descfile.md).

### [--learn-pairs](https://catboost.ai/docs/en/references/training-parameters/input.md#--learn-pairs)

The path to the input file that contains the [pairs description](https://catboost.ai/docs/en/concepts/input-data_pairs-description.md) for the training dataset.

### [--test-pairs](input#--test-pairs)

The path to the input file that contains the [pairs description](https://catboost.ai/docs/en/concepts/input-data_pairs-description.md) for the validation dataset.

### [--learn-group-weights](https://catboost.ai/docs/en/references/training-parameters/input.md#--learn-group-weights)

The path to the input file that contains the weights of groups. Refer to the [Group weights](https://catboost.ai/docs/en/concepts/input-data_group-weights.md) section for format details.

### [--test-group-weights](https://catboost.ai/docs/en/references/training-parameters/input.md#--test-group-weights)

The path to the input file that contains the weights of groups for the validation dataset.

### [--learn-baseline](https://catboost.ai/docs/en/references/training-parameters/input.md#--learn-baseline)

The path to the input file that contains baseline values for the training dataset.

### [--test-baseline](https://catboost.ai/docs/en/references/training-parameters/input.md#--test-baseline)

The path to the input file that contains baseline values for the validation dataset.

### [--delimiter](https://catboost.ai/docs/en/references/training-parameters/input.md#--delimiter)

The delimiter character used to separate the data in the dataset input file.

### [--has-header](input#--has-header)

Read the column names from the first line of the dataset file if this parameter is set.

### [--params-files](https://catboost.ai/docs/en/references/training-parameters/input.md#--params-files)

The path to the input JSON file that contains the training parameters, for example:

### [--nan-mode](https://catboost.ai/docs/en/references/training-parameters/input.md#--nan-mode)

The method for [processing missing values](https://catboost.ai/docs/en/concepts/algorithm-missing-values-processing.md) in the input dataset.

## Multiclassification settings

### [classes_count](https://catboost.ai/docs/en/references/training-parameters/multiclassification.md#classes_count)

Command-line: `--classes-count`

<!-- source: en/_includes/work_src/reusage/classes-count__main-desc.md -->
The upper limit for the numeric class label. Defines the number of classes for multiclassification.
<!-- endsource: en/_includes/work_src/reusage/classes-count__main-desc.md -->

### [--class-names](https://catboost.ai/docs/en/references/training-parameters/multiclassification.md#--class-names)

This parameter is only for Command-line.

Classes names. Allows to redefine the default values when using the MultiClass and Logloss metrics.

## Output settings

### [logging_level](https://catboost.ai/docs/en/references/training-parameters/output.md#logging_level)

Command line: `--logging-level`

The logging level to output to stdout.

### [metric_period](https://catboost.ai/docs/en/references/training-parameters/output.md#metric_period)

Command line: `--metric-period`

The frequency of iterations to calculate the values of [objectives and metrics](https://catboost.ai/docs/en/concepts/loss-functions.md).

The usage of this parameter speeds up the training.

### [verbose](https://catboost.ai/docs/en/references/training-parameters/output.md#verbose)

Command line: `--verbose`

_Alias:_`verbose_eval`

<!-- source: en/_includes/work_src/reusage/python__feature-importances__verbose__short-description__list-intro.md -->
The purpose of this parameter depends on the type of the given value:
<!-- endsource: en/_includes/work_src/reusage/python__feature-importances__verbose__short-description__list-intro.md -->

### [train_dir](https://catboost.ai/docs/en/references/training-parameters/output.md#train_dir)

Command line: `--train-dir`

The directory for storing the files generated during training.

### [model_size_reg](https://catboost.ai/docs/en/references/training-parameters/output.md#model_size_reg)

Command line: `--model-size-reg`

The model size regularization coefficient. The larger the value, the smaller the model size. Refer to the [Model size regularization coefficient](https://catboost.ai/docs/en/references/model-size-reg.md) section for details.

This regularization is needed only for models with categorical features (other models are small).

### [allow_writing_files](https://catboost.ai/docs/en/references/training-parameters/output.md#allow_writing_files)

Allow to write analytical and snapshot files during training.

### [save_snapshot](https://catboost.ai/docs/en/references/training-parameters/output.md#save_snapshot)

Enable snapshotting for [restoring the training progress after an interruption](https://catboost.ai/docs/en/features/snapshots.md).

### [snapshot_file](https://catboost.ai/docs/en/references/training-parameters/output.md#snapshot_file)

The name of the file to save the training progress information in. This file is used for [recovering training after an interruption](https://catboost.ai/docs/en/features/snapshots.md).

### [snapshot_interval](https://catboost.ai/docs/en/references/training-parameters/output.md#snapshot_interval)

The interval between saving snapshots in seconds.

### [roc_file](https://catboost.ai/docs/en/references/training-parameters/output.md#roc_file)

The name of the [output file](https://catboost.ai/docs/en/concepts/output-data_roc-curve-points.md) to save the ROC curve points to.

## Overfitting detection settings

### [early_stopping_rounds](https://catboost.ai/docs/en/references/training-parameters/overfitting-detection.md#early_stopping_rounds)

Sets the overfitting detector type to Iter and stops the training after the specified number of iterations since the iteration with the optimal metric value.

### [od_type](https://catboost.ai/docs/en/references/training-parameters/overfitting-detection.md#od_type)

Command-line: `--od-type`

The type of the overfitting detector to use.

### [od_pval](https://catboost.ai/docs/en/references/training-parameters/overfitting-detection.md#od_pval)

Command-line: `--od-pval`

The threshold for the IncToDec [overfitting detector](https://catboost.ai/docs/en/concepts/overfitting-detector.md) type.

### [od_wait](https://catboost.ai/docs/en/references/training-parameters/overfitting-detection.md#od_wait)

Command-line: `--od-wait`

The number of iterations to continue the training after the iteration with the optimal metric value.

## Performance settings

### [thread_count](https://catboost.ai/docs/en/references/training-parameters/performance.md#thread_count)

Command-line: `-T`, `--thread-count`

The number of threads to use during the training.

### [used_ram_limit](https://catboost.ai/docs/en/references/training-parameters/performance.md#used_ram_limit)

Command-line: `--used-ram-limit`

Attempt to limit the amount of used CPU RAM.

### [gpu_ram_part](https://catboost.ai/docs/en/references/training-parameters/performance.md#gpu_ram_part)

Command-line: `--gpu-ram-part`

How much of the GPU RAM to use for training.

### [pinned_memory_size](https://catboost.ai/docs/en/references/training-parameters/performance.md#pinned_memory_size)

Command-line: `--pinned-memory-size`

How much pinned (page-locked) CPU RAM to use per GPU.

### [gpu_cat_features_storage](https://catboost.ai/docs/en/references/training-parameters/performance.md#gpu_cat_features_storage)

Command-line: `--gpu-cat-features-storage`

The method for storing the categorical features' values.

### [data_partition](https://catboost.ai/docs/en/references/training-parameters/performance.md#data_partition)

Command-line: `--data-partition`

The method for splitting the input dataset between multiple workers.

## Processing unit settings

### [task_type](https://catboost.ai/docs/en/references/training-parameters/processing-unit.md#task_type)

Command line: `--task-type`

The processing unit type to use for training.

### [devices](#devices)

Command line: `--devices`

IDs of the GPU devices to use for training (indices are zero-based).

## Quantization settings

### [target_border](https://catboost.ai/docs/en/references/training-parameters/quantization.md#target_border)

Command-line: `--target-border`

If set, defines the border for converting target values to 0 and 1.

### [border_count](https://catboost.ai/docs/en/references/training-parameters/quantization.md#border_count)

Command-line: `-x`, `--border-count`

_Alias:_ `max_bin`

The number of splits for numerical features. Allowed values are integers from 1 to 65535 inclusively.

### [feature_border_type](https://catboost.ai/docs/en/references/training-parameters/quantization.md#feature_border_type)

Command-line: `--feature-border-type`

The [quantization mode](https://catboost.ai/docs/en/concepts/quantization.md) for numerical features.

### [per_float_feature_quantization](https://catboost.ai/docs/en/references/training-parameters/quantization.md#per_float_feature_quantization)

Command-line: `--per-float-feature-quantization`

The quantization description for the specified feature or list of features.

## Text processing parameters

These parameters are only for the Python package and Command-line version.

### [tokenizers](https://catboost.ai/docs/en/references/training-parameters/text-processing.md#tokenizers)

Command-line: `--tokenizers`

Tokenizers used to preprocess Text type feature columns before creating the dictionary.

### [dictionaries](https://catboost.ai/docs/en/references/training-parameters/text-processing.md#dictionaries)

Command-line: `--dictionaries`

<!-- source: en/_includes/work_src/reusage/cli__dictionaries__desc__div.md -->
Dictionaries used to preprocess Text type feature columns.

Format:
<!-- endsource: en/_includes/work_src/reusage/cli__dictionaries__desc__div.md -->

### [feature_calcers](https://catboost.ai/docs/en/references/training-parameters/text-processing.md#feature_calcers)

Command-line: `--feature-calcers`

<!-- source: en/_includes/work_src/reusage/cli__feature-calcers__desc__div.md -->
Feature calcers used to calculate new features based on preprocessed Text type feature columns.

Format:
<!-- endsource: en/_includes/work_src/reusage/cli__feature-calcers__desc__div.md -->

### [text_processing](https://catboost.ai/docs/en/references/training-parameters/text-processing.md#text_processing)

Command-line: `--text-processing`

<!-- source: en/_includes/work_src/reusage/cli__text-processing__div.md -->
A JSON specification of tokenizers, dictionaries and feature calcers, which determine how text features are converted into a list of float features.

[Example](https://catboost.ai/docs/en/references/text-processing__specification-example.md)

Refer to the description of the following parameters for details on supported values:
<!-- endsource: en/_includes/work_src/reusage/cli__text-processing__div.md -->

## Visualization settings

These parameters are only for the Python package.

### [name](https://catboost.ai/docs/en/references/training-parameters/visualization.md#name)

The experiment name to display in [visualization tools](https://catboost.ai/docs/en/features/visualization.md).