🕷️ Crawler Inspector

URL Lookup

Direct Parameter Lookup

Raw Queries and Responses

1. Shard Calculation

Query:

Response:

Calculated Shard: 169 (from laksa128)

2. Crawled Status Check

Query:

curl -X POST \
  'http://laksa169.int.ahrefs:8124/' \
  -H 'Content-Type: text/plain' \
  -H 'X-ClickHouse-Database: crawler3' \
  -H 'Authorization: Basic YXBpOg==' \
  -d 'SELECT getAhrefsURLFromUnparsed(src_unparsed) AS found_url, ifNull(toUnixTimestamp(download_stamp), 0) AS crawl_time, ifNull(toUnixTimestamp(props_url_first_seen), 0) AS first_indexed_time, download_http_code AS http_code, src_unparsed AS src_unparsed, src_root_hash AS src_root_hash, history_drop_reason AS history_drop_reason, meta_title AS meta_title, meta_descriptions AS meta_descriptions, attrs_boilerpipe_text AS attrs_boilerpipe_text, attrs_markdown AS attrs_markdown, attrs_readable_markdown AS attrs_readable_markdown, meta_canonical AS meta_canonical FROM crawler3.page_info_local FINAL PREWHERE (src_root_hash, src_unparsed) IN ((getAhrefsRootHashFromUnparsed(getAhrefsUnparsedNoserviceFromURL(\'https://catboost.ai/docs/en/features/loss-functions-desc\')), getAhrefsUnparsedNoserviceFromURL(\'https://catboost.ai/docs/en/features/loss-functions-desc\'))) FORMAT JSONEachRow'

Response:

{"found_url":"https:\/\/catboost.ai\/docs\/en\/features\/loss-functions-desc","crawl_time":1775529257,"first_indexed_time":1732070880,"http_code":200,"src_unparsed":"ai,catboost!\/docs\/en\/features\/loss-functions-desc s443","src_root_hash":"17435841955170310369","history_drop_reason":null,"meta_title":"Implemented metrics | CatBoost","meta_descriptions":["CatBoost provides built-in metrics for various machine learning problems. These functions can be used for model optimization or reference purposes. See the Obje"],"attrs_boilerpipe_text":"CatBoost provides built-in metrics for various machine learning problems. These functions can be used for model optimization or reference purposes. See the \nObjectives and metrics\nsection for details on the calculation principles.\nChoose the implementation for more details.\npython\nr-package\ncli\nPython package\nThe following parameters can be set for the corresponding classes and are used when the model is trained.\nParameters for trained model\nClasses:\nCatBoost\nCatBoostClassifier\nCatBoostRegressor\nloss-function\nThe\nmetric\nto use in training. The specified value also determines the machine learning problem to solve. Some metrics support optional parameters (see the \nObjectives and metrics\nsection for details on each metric).\nFormat:\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\nSupported metrics\nRMSE\nLogloss\nMAE\nCrossEntropy\nQuantile\nLogLinQuantile\nLq\nMultiRMSE\nMultiClass\nMultiClassOneVsAll\nMultiLogloss\nMultiCrossEntropy\nMAPE\nPoisson\nPairLogit\nPairLogitPairwise\nQueryRMSE\nQuerySoftMax\nGroupQuantile\nTweedie\nYetiRank\nYetiRankPairwise\nStochasticFilter\nStochasticRank\nA custom python object can also be set as the value of this parameter (see an \nexample\n).\nFor example, use the following construction to calculate the value of Quantile with the coefficient \nα\n=\n0.1\n\\alpha = 0.1\n:\nQuantile:alpha=0.1\ncustom_metric\nMetric\nvalues to output during training. These functions are not optimized and are displayed for informational purposes only. Some metrics support optional parameters (see the \nObjectives and metrics\nsection for details on each metric).\nFormat:\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\nSupported metrics\nExamples:\nCalculate the value of CrossEntropy\nCrossEntropy\nCalculate the value of Quantile with the coefficient \nα\n=\n0.1\n\\alpha = 0.1\nQuantile:alpha=0.1\nCalculate the values of Logloss and AUC\n[\n'Logloss'\n,\n'AUC'\n]\nValues of all custom metrics for learn and validation datasets are saved to the \nMetric\noutput files (\nlearn_error.tsv\nand\ntest_error.tsv\nrespectively). The directory for these files is specified in the \n--train-dir\n(\ntrain_dir\n) parameter.\nUse the \nvisualization tools\nto see a live chart with the dynamics of the specified metrics.\nuse-best-model\nIf this parameter is set, the number of trees that are saved in the resulting model is defined as follows:\nBuild the number of trees defined by the training parameters.\nUse the validation dataset to identify the iteration with the optimal value of the metric specified in  \n--eval-metric\n(\n--eval-metric\n).\nNo trees are saved after this iteration.\nThis option requires a validation dataset to be provided.\neval-metric\nThe metric used for overfitting detection (if enabled) and best model selection (if enabled). Some metrics support optional parameters (see the \nObjectives and metrics\nsection for details on each metric).\nFormat:\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\nSupported metrics\nA user-defined function can also be set as the value (see an \nexample\n).\nExamples:\nR2\nThe following parameters can be set for the corresponding methods and are used when the model is trained or applied.\nParameters for trained or applied model\nThe following parameters can be set for the corresponding methods and are used when the model is trained or applied.\nClasses:\nfit\n(\nCatBoost\n)\nfit\n(\nCatBoostClassifier\n)\nfit\n(\nCatBoostRegressor\n)\nuse_best_model\nIf this parameter is set, the number of trees that are saved in the resulting model is defined as follows:\nBuild the number of trees defined by the training parameters.\nUse the validation dataset to identify the iteration with the optimal value of the metric specified in  \n--eval-metric\n(\n--eval-metric\n).\nNo trees are saved after this iteration.\nThis option requires a validation dataset to be provided.\nverbose\nOutput the measured evaluation metric to stderr.\nplot\nPlot the following information during training:\nthe metric values;\nthe custom loss values;\nthe loss function change during feature selection;\nthe time has passed since training started;\nthe remaining time until the end of training.\nThis \noption can be used\nif training is performed in Jupyter notebook.\nR package\nThe following parameters can be set for the corresponding methods and are used when the model is trained or applied.\nMethod:\ncatboost.train\nloss_function\nDescription\nThe\nmetric\nto use in training. The specified value also determines the machine learning problem to solve. Some metrics support optional parameters (see the \nObjectives and metrics\nsection for details on each metric).\nFormat:\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\nSupported metrics\nRMSE\nLogloss\nMAE\nCrossEntropy\nQuantile\nLogLinQuantile\nLq\nMultiRMSE\nMultiClass\nMultiClassOneVsAll\nMultiLogloss\nMultiCrossEntropy\nMAPE\nPoisson\nPairLogit\nPairLogitPairwise\nQueryRMSE\nQuerySoftMax\nGroupQuantile\nTweedie\nYetiRank\nYetiRankPairwise\nStochasticFilter\nStochasticRank\nFor example, use the following construction to calculate the value of Quantile with the coefficient \nα\n=\n0.1\n\\alpha = 0.1\n:\nQuantile:alpha=0.1\ncustom_loss\nParameters\nMetric\nvalues to output during training. These functions are not optimized and are displayed for informational purposes only. Some metrics support optional parameters (see the \nObjectives and metrics\nsection for details on each metric).\nFormat:\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\nSupported metrics\nExamples:\nCalculate the value of CrossEntropy\nc('CrossEntropy')\nOr simply:\n'CrossEntropy'\nCalculate the values of Logloss and AUC\nc('Logloss', 'AUC')\nCalculate the value of Quantile with the coefficient \nα\n=\n0.1\n\\alpha = 0.1\nc('Quantilealpha=0.1')\nValues of all custom metrics for learn and validation datasets are saved to the \nMetric\noutput files (\nlearn_error.tsv\nand\ntest_error.tsv\nrespectively). The directory for these files is specified in the \n--train-dir\n(\ntrain_dir\n) parameter.\nuse-best-model\nIf this parameter is set, the number of trees that are saved in the resulting model is defined as follows:\nBuild the number of trees defined by the training parameters.\nUse the validation dataset to identify the iteration with the optimal value of the metric specified in  \n--eval-metric\n(\n--eval-metric\n).\nNo trees are saved after this iteration.\nThis option requires a validation dataset to be provided.\neval-metric\nParameters\nThe metric used for overfitting detection (if enabled) and best model selection (if enabled). Some metrics support optional parameters (see the \nObjectives and metrics\nsection for details on each metric).\nFormat:\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\nSupported metrics\nQuantile:alpha=0.3\nCommand-line version\nThe following command keys can be specified for the corresponding commands and are used when the model is trained or applied.\nParams for the\ncatboost fit\ncommand:\n--loss-function\nThe\nmetric\nto use in training. The specified value also determines the machine learning problem to solve. Some metrics support optional parameters (see the \nObjectives and metrics\nsection for details on each metric).\nFormat:\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\nSupported metrics\nRMSE\nLogloss\nMAE\nCrossEntropy\nQuantile\nLogLinQuantile\nLq\nMultiRMSE\nMultiClass\nMultiClassOneVsAll\nMultiLogloss\nMultiCrossEntropy\nMAPE\nPoisson\nPairLogit\nPairLogitPairwise\nQueryRMSE\nQuerySoftMax\nGroupQuantile\nTweedie\nYetiRank\nYetiRankPairwise\nStochasticFilter\nStochasticRank\nFor example, use the following construction to calculate the value of Quantile with the coefficient \nα\n=\n0.1\n\\alpha = 0.1\n:\nQuantilealpha=0.1\n--custom-metric\nMetric\nvalues to output during training. These functions are not optimized and are displayed for informational purposes only. Some metrics support optional parameters (see the \nObjectives and metrics\nsection for details on each metric).\nFormat:\n<Metric 1>[:<parameter 1>=<value>;..;<parameter N>=<value>],<Metric 2>[:<parameter 1>=<value>;..;<parameter N>=<value>],..,<Metric N>[:<parameter 1>=<value>;..;<parameter N>=<value>]\nSupported metrics\nExamples:\nCalculate the value of CrossEntropy\nCrossEntropy\nCalculate the value of Quantile with the coefficient \nα\n=\n0.1\n\\alpha = 0.1\nQuantilealpha=0.1\nValues of all custom metrics for learn and validation datasets are saved to the \nMetric\noutput files (\nlearn_error.tsv\nand\ntest_error.tsv\nrespectively). The directory for these files is specified in the \n--train-dir\n(\ntrain_dir\n) parameter.\n--use-best-model\nIf this parameter is set, the number of trees that are saved in the resulting model is defined as follows:\nBuild the number of trees defined by the training parameters.\nUse the validation dataset to identify the iteration with the optimal value of the metric specified in  \n--eval-metric\n(\n--eval-metric\n).\nNo trees are saved after this iteration.\nThis option requires a validation dataset to be provided.\n--eval-metric\nThe metric used for overfitting detection (if enabled) and best model selection (if enabled). Some metrics support optional parameters (see the \nObjectives and metrics\nsection for details on each metric).\nFormat:\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\nSupported metrics\nExamples:\nR2\nQuantile:alpha=0.3\n--logging-level\nThe logging level to output to stdout.\nPossible values:\nSilent — Do not output any logging information to stdout.\nVerbose — Output the following data to stdout:\noptimized metric\nelapsed time of training\nremaining time of training\nInfo — Output additional information and the number of trees.\nDebug — Output debugging information.","attrs_markdown":"[![Logo icon](https:\/\/yastatic.net\/s3\/locdoc\/daas-static\/catboost\/71b237a322eec6f2889af0dae2a9c549.svg)](https:\/\/catboost.ai\/ \"CatBoost\")\n\n- Installation\n  - [Overview](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/installation)\n  - Python package installation\n  - CatBoost for Apache Spark installation\n  - R package installation\n  - Command-line version binary\n  - Build from source\n- Key Features\n  - [Training](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/training)\n  - [Training on GPU](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/training-on-gpu)\n  - [Regular prediction](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/prediction)\n  - [Staged prediction](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/staged-prediction)\n  - [Cross-validation](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/cross-validation)\n  - [Feature importances](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/feature-importances-calculation)\n  - [User-defined metrics](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/custom-loss-functions)\n  - [Using the overfitting detector](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/overfitting-detector-desc)\n  - [Export a model to CoreML](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/export-model-to-core-ml)\n  - [Pre-trained data](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/proceed-training)\n  - [Calculate metrics](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/eval-metrics)\n  - [Categorical features](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/categorical-features)\n  - [Text features](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/text-features)\n  - [Embeddings features](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/embeddings-features)\n  - [Aggregated graph features](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/graph-aggregated-features)\n  - [Implemented metrics](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc)\n  - [Export a model to Python or C++](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/export-model-to-python)\n  - [Export a model to JSON](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/export-model-to-json)\n  - [Object importances](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/object-importances-calcution)\n- Training parameters\n- Python package\n- CatBoost for Apache Spark\n- R package\n- Command-line version\n- Applying models\n- Objectives and metrics\n- Model analysis\n- Data format description\n- [Parameter tuning](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/parameter-tuning)\n- [Speeding up the training](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/speed-up-training)\n- Data visualization\n- Algorithm details\n- [FAQ](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/faq)\n- Educational materials\n- [Development and contributions](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/development-and-contributions)\n- [Contacts](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/contacts)\n\nImplemented metrics\n\n## In this article:\n- [Python package](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#python-package)\n- [Parameters for trained model](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#parameters-for-trained-model)\n- [Parameters for trained or applied model](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#parameters-for-trained-or-applied-model)\n- [R package](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#r-package)\n- [loss\\_function](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#loss_function)\n- [custom\\_loss](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#custom_loss)\n- [use-best-model](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#use-best-model1)\n- [eval-metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#eval-metric1)\n- [Command-line version](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#command-line-version)\n- [\\--loss-function](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#loss-function1)\n- [\\--custom-metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#custom-metric)\n- [\\--use-best-model](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#use-best-model2)\n- [\\--eval-metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#eval-metric2)\n- [\\--logging-level](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#logging-level)\n\n1. Key Features\n2. Implemented metrics\n\n# Implemented metrics\n- [Python package](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#python-package)\n  - [Parameters for trained model](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#parameters-for-trained-model)\n  - [Parameters for trained or applied model](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#parameters-for-trained-or-applied-model)\n- [R package](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#r-package)\n  - [loss\\_function](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#loss_function)\n  - [custom\\_loss](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#custom_loss)\n  - [use-best-model](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#use-best-model1)\n  - [eval-metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#eval-metric1)\n- [Command-line version](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#command-line-version)\n  - [\\--loss-function](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#loss-function1)\n  - [\\--custom-metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#custom-metric)\n  - [\\--use-best-model](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#use-best-model2)\n  - [\\--eval-metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#eval-metric2)\n  - [\\--logging-level](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#logging-level)\n\nCatBoost provides built-in metrics for various machine learning problems. These functions can be used for model optimization or reference purposes. See the [Objectives and metrics](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/loss-functions) section for details on the calculation principles.\n\nChoose the implementation for more details.\n\n- [python](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#python)\n- [r-package](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#r-package)\n- [cli](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/loss-functions-desc#command-line-version)\n## Python package\nThe following parameters can be set for the corresponding classes and are used when the model is trained.\n\n### Parameters for trained model\nClasses:\n\n- [CatBoost](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/python-reference_catboost)\n- [CatBoostClassifier](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/python-reference_catboostclassifier)\n- [CatBoostRegressor](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/python-reference_catboostregressor)\n#### loss-function\nThe [metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/loss-functions) 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\/features\/en\/concepts\/loss-functions) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\nSupported metrics\n\n- RMSE\n- Logloss\n- MAE\n- CrossEntropy\n- Quantile\n- LogLinQuantile\n- Lq\n- MultiRMSE\n- MultiClass\n- MultiClassOneVsAll\n- MultiLogloss\n- MultiCrossEntropy\n- MAPE\n- Poisson\n- PairLogit\n- PairLogitPairwise\n- QueryRMSE\n- QuerySoftMax\n- GroupQuantile\n- Tweedie\n- YetiRank\n- YetiRankPairwise\n- StochasticFilter\n- StochasticRank\n\nA custom python object can also be set as the value of this parameter (see an [example](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/python-usages-examples)).\n\nFor example, use the following construction to calculate the value of Quantile with the coefficient α \\= 0\\.1 \\\\alpha = 0.1 α\\=0\\.1:\n```\nQuantile:alpha=0.1\n```\n#### custom\\_metric\n[Metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/loss-functions) 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\/features\/en\/concepts\/loss-functions) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\n[Supported metrics](https:\/\/catboost.ai\/docs\/en\/features\/en\/references\/custom-metric__supported-metrics)\n\nExamples:\n\n- Calculate the value of CrossEntropy\n  ```\n  CrossEntropy\n  ```\n- Calculate the value of Quantile with the coefficient α \\= 0\\.1 \\\\alpha = 0.1 α\\=0\\.1\n  ```\n  Quantile:alpha=0.1\n  ```\n- Calculate the values of Logloss and AUC\n  ```\n  ['Logloss', 'AUC']\n  ```\n\nValues of all custom metrics for learn and validation datasets are saved to the [Metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/output-data_loss-function) output files (`learn_error.tsv` and `test_error.tsv` respectively). The directory for these files is specified in the `--train-dir` (`train_dir`) parameter.\n\nUse the [visualization tools](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/visualization) to see a live chart with the dynamics of the specified metrics.\n\n#### use-best-model\nIf this parameter is set, the number of trees that are saved in the resulting model is defined as follows:\n\n1. Build the number of trees defined by the training parameters.\n2. Use the validation dataset to identify the iteration with the optimal value of the metric specified in `--eval-metric` (`--eval-metric`).\n\nNo trees are saved after this iteration.\n\nThis option requires a validation dataset to be provided.\n\n#### eval-metric\nThe 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\/features\/en\/concepts\/loss-functions) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\n[Supported metrics](https:\/\/catboost.ai\/docs\/en\/features\/en\/references\/eval-metric__supported-metrics)\n\nA user-defined function can also be set as the value (see an [example](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/python-usages-examples)).\n\nExamples:\n```\nR2\n```\nThe following parameters can be set for the corresponding methods and are used when the model is trained or applied.\n\n### Parameters for trained or applied model\nThe following parameters can be set for the corresponding methods and are used when the model is trained or applied.\n\nClasses:\n\n- [fit](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/python-reference_catboost_fit) ([CatBoost](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/python-reference_catboost))\n- [fit](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/python-reference_catboostclassifier_fit) ([CatBoostClassifier](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/python-reference_catboostclassifier))\n- [fit](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/python-reference_catboostregressor_fit) ([CatBoostRegressor](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/python-reference_catboostregressor))\n#### use\\_best\\_model\nIf this parameter is set, the number of trees that are saved in the resulting model is defined as follows:\n\n1. Build the number of trees defined by the training parameters.\n2. Use the validation dataset to identify the iteration with the optimal value of the metric specified in `--eval-metric` (`--eval-metric`).\n\nNo trees are saved after this iteration.\n\nThis option requires a validation dataset to be provided.\n\n#### verbose\nOutput the measured evaluation metric to stderr.\n\n#### plot\nPlot the following information during training:\n\n- the metric values;\n- the custom loss values;\n- the loss function change during feature selection;\n- the time has passed since training started;\n- the remaining time until the end of training.  \n   This [option can be used](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/visualization_jupyter-notebook) if training is performed in Jupyter notebook.\n## R package\nThe following parameters can be set for the corresponding methods and are used when the model is trained or applied.\n\nMethod: [catboost.train](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/r-reference_catboost-train)\n\n### loss\\_function\n**Description**\n\nThe [metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/loss-functions) 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\/features\/en\/concepts\/loss-functions) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\nSupported metrics\n\n- RMSE\n- Logloss\n- MAE\n- CrossEntropy\n- Quantile\n- LogLinQuantile\n- Lq\n- MultiRMSE\n- MultiClass\n- MultiClassOneVsAll\n- MultiLogloss\n- MultiCrossEntropy\n- MAPE\n- Poisson\n- PairLogit\n- PairLogitPairwise\n- QueryRMSE\n- QuerySoftMax\n- GroupQuantile\n- Tweedie\n- YetiRank\n- YetiRankPairwise\n- StochasticFilter\n- StochasticRank\n\nFor example, use the following construction to calculate the value of Quantile with the coefficient α \\= 0\\.1 \\\\alpha = 0.1 α\\=0\\.1:\n```\nQuantile:alpha=0.1\n```\n### custom\\_loss\n**Parameters**\n\n[Metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/loss-functions) 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\/features\/en\/concepts\/loss-functions) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\n[Supported metrics](https:\/\/catboost.ai\/docs\/en\/features\/en\/references\/custom-metric__supported-metrics)\n\nExamples:\n\n- Calculate the value of CrossEntropy\n  ```\n  c('CrossEntropy')\n  ```\n  Or simply:\n  ```\n  'CrossEntropy'\n  ```\n- Calculate the values of Logloss and AUC\n  ```\n  c('Logloss', 'AUC')\n  ```\n- Calculate the value of Quantile with the coefficient α \\= 0\\.1 \\\\alpha = 0.1 α\\=0\\.1\n  ```\n  c('Quantilealpha=0.1')\n  ```\n\nValues of all custom metrics for learn and validation datasets are saved to the [Metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/output-data_loss-function) output files (`learn_error.tsv` and `test_error.tsv` respectively). The directory for these files is specified in the `--train-dir` (`train_dir`) parameter.\n\n### use-best-model\nIf this parameter is set, the number of trees that are saved in the resulting model is defined as follows:\n\n1. Build the number of trees defined by the training parameters.\n2. Use the validation dataset to identify the iteration with the optimal value of the metric specified in `--eval-metric` (`--eval-metric`).\n\nNo trees are saved after this iteration.\n\nThis option requires a validation dataset to be provided.\n\n### eval-metric\n**Parameters**\n\nThe 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\/features\/en\/concepts\/loss-functions) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\n[Supported metrics](https:\/\/catboost.ai\/docs\/en\/features\/en\/references\/eval-metric__supported-metrics)\n```\nQuantile:alpha=0.3\n```\n## Command-line version\nThe following command keys can be specified for the corresponding commands and are used when the model is trained or applied.\n\nParams for the [catboost fit](https:\/\/catboost.ai\/docs\/en\/features\/en\/references\/training-parameters\/) command:\n\n### \\--loss-function\nThe [metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/loss-functions) 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\/features\/en\/concepts\/loss-functions) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\nSupported metrics\n\n- RMSE\n- Logloss\n- MAE\n- CrossEntropy\n- Quantile\n- LogLinQuantile\n- Lq\n- MultiRMSE\n- MultiClass\n- MultiClassOneVsAll\n- MultiLogloss\n- MultiCrossEntropy\n- MAPE\n- Poisson\n- PairLogit\n- PairLogitPairwise\n- QueryRMSE\n- QuerySoftMax\n- GroupQuantile\n- Tweedie\n- YetiRank\n- YetiRankPairwise\n- StochasticFilter\n- StochasticRank\n\nFor example, use the following construction to calculate the value of Quantile with the coefficient α \\= 0\\.1 \\\\alpha = 0.1 α\\=0\\.1:\n```\nQuantilealpha=0.1\n```\n### \\--custom-metric\n[Metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/loss-functions) 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\/features\/en\/concepts\/loss-functions) section for details on each metric).\n\nFormat:\n```\n<Metric 1>[:<parameter 1>=<value>;..;<parameter N>=<value>],<Metric 2>[:<parameter 1>=<value>;..;<parameter N>=<value>],..,<Metric N>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\n[Supported metrics](https:\/\/catboost.ai\/docs\/en\/features\/en\/references\/custom-metric__supported-metrics)\n\nExamples:\n\n- Calculate the value of CrossEntropy\n  ```\n  CrossEntropy\n  ```\n- Calculate the value of Quantile with the coefficient α \\= 0\\.1 \\\\alpha = 0.1 α\\=0\\.1\n  ```\n  Quantilealpha=0.1\n  ```\n\nValues of all custom metrics for learn and validation datasets are saved to the [Metric](https:\/\/catboost.ai\/docs\/en\/features\/en\/concepts\/output-data_loss-function) output files (`learn_error.tsv` and `test_error.tsv` respectively). The directory for these files is specified in the `--train-dir` (`train_dir`) parameter.\n\n### \\--use-best-model\nIf this parameter is set, the number of trees that are saved in the resulting model is defined as follows:\n\n1. Build the number of trees defined by the training parameters.\n2. Use the validation dataset to identify the iteration with the optimal value of the metric specified in `--eval-metric` (`--eval-metric`).\n\nNo trees are saved after this iteration.\n\nThis option requires a validation dataset to be provided.\n\n### \\--eval-metric\nThe 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\/features\/en\/concepts\/loss-functions) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\n[Supported metrics](https:\/\/catboost.ai\/docs\/en\/features\/en\/references\/eval-metric__supported-metrics)\n\nExamples:\n```\nR2\n```\n```\nQuantile:alpha=0.3\n```\n### \\--logging-level\nThe logging level to output to stdout.\n\nPossible values:\n\n- Silent — Do not output any logging information to stdout.\n- Verbose — Output the following data to stdout:\n  - optimized metric\n  - elapsed time of training\n  - remaining time of training\n- Info — Output additional information and the number of trees.\n- Debug — Output debugging information.\n\n### Was the article helpful?\nYes\n\nNo\n\nPrevious\n\n[Aggregated graph features](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/graph-aggregated-features)\n\nNext\n\n[Export a model to Python or C++](https:\/\/catboost.ai\/docs\/en\/features\/en\/features\/export-model-to-python)\n\n![](https:\/\/mc.yandex.ru\/watch\/60763294)","attrs_readable_markdown":"CatBoost provides built-in metrics for various machine learning problems. These functions can be used for model optimization or reference purposes. See the [Objectives and metrics](https:\/\/catboost.ai\/docs\/en\/concepts\/loss-functions) section for details on the calculation principles.\n\nChoose the implementation for more details.\n\n- [python](https:\/\/catboost.ai\/docs\/en\/features\/loss-functions-desc#python)\n- [r-package](https:\/\/catboost.ai\/docs\/en\/features\/loss-functions-desc#r-package)\n- [cli](https:\/\/catboost.ai\/docs\/en\/features\/loss-functions-desc#command-line-version)\n## Python package\nThe following parameters can be set for the corresponding classes and are used when the model is trained.\n\n### Parameters for trained model\nClasses:\n\n- [CatBoost](https:\/\/catboost.ai\/docs\/en\/concepts\/python-reference_catboost)\n- [CatBoostClassifier](https:\/\/catboost.ai\/docs\/en\/concepts\/python-reference_catboostclassifier)\n- [CatBoostRegressor](https:\/\/catboost.ai\/docs\/en\/concepts\/python-reference_catboostregressor)\n#### loss-function\nThe [metric](https:\/\/catboost.ai\/docs\/en\/concepts\/loss-functions) 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) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\nSupported metrics\n\n- RMSE\n- Logloss\n- MAE\n- CrossEntropy\n- Quantile\n- LogLinQuantile\n- Lq\n- MultiRMSE\n- MultiClass\n- MultiClassOneVsAll\n- MultiLogloss\n- MultiCrossEntropy\n- MAPE\n- Poisson\n- PairLogit\n- PairLogitPairwise\n- QueryRMSE\n- QuerySoftMax\n- GroupQuantile\n- Tweedie\n- YetiRank\n- YetiRankPairwise\n- StochasticFilter\n- StochasticRank\n\nA custom python object can also be set as the value of this parameter (see an [example](https:\/\/catboost.ai\/docs\/en\/concepts\/python-usages-examples)).\n\nFor example, use the following construction to calculate the value of Quantile with the coefficient α \\= 0\\.1 \\\\alpha = 0.1:\n```\nQuantile:alpha=0.1\n```\n#### custom\\_metric\n[Metric](https:\/\/catboost.ai\/docs\/en\/concepts\/loss-functions) 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) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\n[Supported metrics](https:\/\/catboost.ai\/docs\/en\/references\/custom-metric__supported-metrics)\n\nExamples:\n\n- Calculate the value of CrossEntropy\n  ```\n  CrossEntropy\n  ```\n- Calculate the value of Quantile with the coefficient α \\= 0\\.1 \\\\alpha = 0.1\n  ```\n  Quantile:alpha=0.1\n  ```\n- Calculate the values of Logloss and AUC\n  ```\n  ['Logloss', 'AUC']\n  ```\n\nValues of all custom metrics for learn and validation datasets are saved to the [Metric](https:\/\/catboost.ai\/docs\/en\/concepts\/output-data_loss-function) output files (`learn_error.tsv` and `test_error.tsv` respectively). The directory for these files is specified in the `--train-dir` (`train_dir`) parameter.\n\nUse the [visualization tools](https:\/\/catboost.ai\/docs\/en\/features\/visualization) to see a live chart with the dynamics of the specified metrics.\n\n#### use-best-model\nIf this parameter is set, the number of trees that are saved in the resulting model is defined as follows:\n\n1. Build the number of trees defined by the training parameters.\n2. Use the validation dataset to identify the iteration with the optimal value of the metric specified in `--eval-metric` (`--eval-metric`).\n\nNo trees are saved after this iteration.\n\nThis option requires a validation dataset to be provided.\n\n#### eval-metric\nThe 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) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\n[Supported metrics](https:\/\/catboost.ai\/docs\/en\/references\/eval-metric__supported-metrics)\n\nA user-defined function can also be set as the value (see an [example](https:\/\/catboost.ai\/docs\/en\/concepts\/python-usages-examples)).\n\nExamples:\n```\nR2\n```\nThe following parameters can be set for the corresponding methods and are used when the model is trained or applied.\n\n### Parameters for trained or applied model\nThe following parameters can be set for the corresponding methods and are used when the model is trained or applied.\n\nClasses:\n\n- [fit](https:\/\/catboost.ai\/docs\/en\/concepts\/python-reference_catboost_fit) ([CatBoost](https:\/\/catboost.ai\/docs\/en\/concepts\/python-reference_catboost))\n- [fit](https:\/\/catboost.ai\/docs\/en\/concepts\/python-reference_catboostclassifier_fit) ([CatBoostClassifier](https:\/\/catboost.ai\/docs\/en\/concepts\/python-reference_catboostclassifier))\n- [fit](https:\/\/catboost.ai\/docs\/en\/concepts\/python-reference_catboostregressor_fit) ([CatBoostRegressor](https:\/\/catboost.ai\/docs\/en\/concepts\/python-reference_catboostregressor))\n#### use\\_best\\_model\nIf this parameter is set, the number of trees that are saved in the resulting model is defined as follows:\n\n1. Build the number of trees defined by the training parameters.\n2. Use the validation dataset to identify the iteration with the optimal value of the metric specified in `--eval-metric` (`--eval-metric`).\n\nNo trees are saved after this iteration.\n\nThis option requires a validation dataset to be provided.\n\n#### verbose\nOutput the measured evaluation metric to stderr.\n\n#### plot\nPlot the following information during training:\n\n- the metric values;\n- the custom loss values;\n- the loss function change during feature selection;\n- the time has passed since training started;\n- the remaining time until the end of training.  \n   This [option can be used](https:\/\/catboost.ai\/docs\/en\/features\/visualization_jupyter-notebook) if training is performed in Jupyter notebook.\n## R package\nThe following parameters can be set for the corresponding methods and are used when the model is trained or applied.\n\nMethod: [catboost.train](https:\/\/catboost.ai\/docs\/en\/concepts\/r-reference_catboost-train)\n\n### loss\\_function\n**Description**\n\nThe [metric](https:\/\/catboost.ai\/docs\/en\/concepts\/loss-functions) 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) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\nSupported metrics\n\n- RMSE\n- Logloss\n- MAE\n- CrossEntropy\n- Quantile\n- LogLinQuantile\n- Lq\n- MultiRMSE\n- MultiClass\n- MultiClassOneVsAll\n- MultiLogloss\n- MultiCrossEntropy\n- MAPE\n- Poisson\n- PairLogit\n- PairLogitPairwise\n- QueryRMSE\n- QuerySoftMax\n- GroupQuantile\n- Tweedie\n- YetiRank\n- YetiRankPairwise\n- StochasticFilter\n- StochasticRank\n\nFor example, use the following construction to calculate the value of Quantile with the coefficient α \\= 0\\.1 \\\\alpha = 0.1:\n```\nQuantile:alpha=0.1\n```\n### custom\\_loss\n**Parameters**\n\n[Metric](https:\/\/catboost.ai\/docs\/en\/concepts\/loss-functions) 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) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\n[Supported metrics](https:\/\/catboost.ai\/docs\/en\/references\/custom-metric__supported-metrics)\n\nExamples:\n\n- Calculate the value of CrossEntropy\n  ```\n  c('CrossEntropy')\n  ```\n  Or simply:\n  ```\n  'CrossEntropy'\n  ```\n- Calculate the values of Logloss and AUC\n  ```\n  c('Logloss', 'AUC')\n  ```\n- Calculate the value of Quantile with the coefficient α \\= 0\\.1 \\\\alpha = 0.1\n  ```\n  c('Quantilealpha=0.1')\n  ```\n\nValues of all custom metrics for learn and validation datasets are saved to the [Metric](https:\/\/catboost.ai\/docs\/en\/concepts\/output-data_loss-function) output files (`learn_error.tsv` and `test_error.tsv` respectively). The directory for these files is specified in the `--train-dir` (`train_dir`) parameter.\n\n### use-best-model\nIf this parameter is set, the number of trees that are saved in the resulting model is defined as follows:\n\n1. Build the number of trees defined by the training parameters.\n2. Use the validation dataset to identify the iteration with the optimal value of the metric specified in `--eval-metric` (`--eval-metric`).\n\nNo trees are saved after this iteration.\n\nThis option requires a validation dataset to be provided.\n\n### eval-metric\n**Parameters**\n\nThe 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) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\n[Supported metrics](https:\/\/catboost.ai\/docs\/en\/references\/eval-metric__supported-metrics)\n```\nQuantile:alpha=0.3\n```\n## Command-line version\nThe following command keys can be specified for the corresponding commands and are used when the model is trained or applied.\n\nParams for the [catboost fit](https:\/\/catboost.ai\/docs\/en\/references\/training-parameters\/) command:\n\n### \\--loss-function\nThe [metric](https:\/\/catboost.ai\/docs\/en\/concepts\/loss-functions) 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) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\nSupported metrics\n\n- RMSE\n- Logloss\n- MAE\n- CrossEntropy\n- Quantile\n- LogLinQuantile\n- Lq\n- MultiRMSE\n- MultiClass\n- MultiClassOneVsAll\n- MultiLogloss\n- MultiCrossEntropy\n- MAPE\n- Poisson\n- PairLogit\n- PairLogitPairwise\n- QueryRMSE\n- QuerySoftMax\n- GroupQuantile\n- Tweedie\n- YetiRank\n- YetiRankPairwise\n- StochasticFilter\n- StochasticRank\n\nFor example, use the following construction to calculate the value of Quantile with the coefficient α \\= 0\\.1 \\\\alpha = 0.1:\n```\nQuantilealpha=0.1\n```\n### \\--custom-metric\n[Metric](https:\/\/catboost.ai\/docs\/en\/concepts\/loss-functions) 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) section for details on each metric).\n\nFormat:\n```\n<Metric 1>[:<parameter 1>=<value>;..;<parameter N>=<value>],<Metric 2>[:<parameter 1>=<value>;..;<parameter N>=<value>],..,<Metric N>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\n[Supported metrics](https:\/\/catboost.ai\/docs\/en\/references\/custom-metric__supported-metrics)\n\nExamples:\n\n- Calculate the value of CrossEntropy\n  ```\n  CrossEntropy\n  ```\n- Calculate the value of Quantile with the coefficient α \\= 0\\.1 \\\\alpha = 0.1\n  ```\n  Quantilealpha=0.1\n  ```\n\nValues of all custom metrics for learn and validation datasets are saved to the [Metric](https:\/\/catboost.ai\/docs\/en\/concepts\/output-data_loss-function) output files (`learn_error.tsv` and `test_error.tsv` respectively). The directory for these files is specified in the `--train-dir` (`train_dir`) parameter.\n\n### \\--use-best-model\nIf this parameter is set, the number of trees that are saved in the resulting model is defined as follows:\n\n1. Build the number of trees defined by the training parameters.\n2. Use the validation dataset to identify the iteration with the optimal value of the metric specified in `--eval-metric` (`--eval-metric`).\n\nNo trees are saved after this iteration.\n\nThis option requires a validation dataset to be provided.\n\n### \\--eval-metric\nThe 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) section for details on each metric).\n\nFormat:\n```\n<Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>]\n```\n[Supported metrics](https:\/\/catboost.ai\/docs\/en\/references\/eval-metric__supported-metrics)\n\nExamples:\n```\nR2\n```\n```\nQuantile:alpha=0.3\n```\n### \\--logging-level\nThe logging level to output to stdout.\n\nPossible values:\n\n- Silent — Do not output any logging information to stdout.\n- Verbose — Output the following data to stdout:\n  - optimized metric\n  - elapsed time of training\n  - remaining time of training\n- Info — Output additional information and the number of trees.\n- Debug — Output debugging information.","meta_canonical":null}

3. Robots.txt Check

Query:

Response:

4. Spam/Ban Check

Query:

Response:

5. Seen Status Check

ℹ️ Skipped - page is already crawled

📄

INDEXABLE

✅

CRAWLED

4 days ago

🤖

ROBOTS ALLOWED

Page Info Filters

Filter	Status	Condition	Details
HTTP status	PASS	`download_http_code = 200`	HTTP 200
Age cutoff	PASS	`download_stamp > now() - 6 MONTH`	0.1 months ago
History drop	PASS	`isNull(history_drop_reason)`	No drop reason
Spam/ban	PASS	`fh_dont_index != 1 AND ml_spam_score = 0`	ml_spam_score=0
Canonical	PASS	`meta_canonical IS NULL OR = '' OR = src_unparsed`	Not set

Page Details

Property	Value
URL	https://catboost.ai/docs/en/features/loss-functions-desc
Last Crawled	2026-04-07 02:34:17 (4 days ago)
First Indexed	2024-11-20 02:48:00 (1 year ago)
HTTP Status Code	200
Meta Title	Implemented metrics \| CatBoost
Meta Description	CatBoost provides built-in metrics for various machine learning problems. These functions can be used for model optimization or reference purposes. See the Obje
Meta Canonical	null
Boilerpipe Text	CatBoost provides built-in metrics for various machine learning problems. These functions can be used for model optimization or reference purposes. See the Objectives and metrics section for details on the calculation principles. Choose the implementation for more details. python r-package cli Python package The following parameters can be set for the corresponding classes and are used when the model is trained. Parameters for trained model Classes: CatBoost CatBoostClassifier CatBoostRegressor loss-function The metric 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 section for details on each metric). Format: <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] Supported metrics RMSE Logloss MAE CrossEntropy Quantile LogLinQuantile Lq MultiRMSE MultiClass MultiClassOneVsAll MultiLogloss MultiCrossEntropy MAPE Poisson PairLogit PairLogitPairwise QueryRMSE QuerySoftMax GroupQuantile Tweedie YetiRank YetiRankPairwise StochasticFilter StochasticRank A custom python object can also be set as the value of this parameter (see an example ). For example, use the following construction to calculate the value of Quantile with the coefficient α = 0.1 \alpha = 0.1 : Quantile:alpha=0.1 custom_metric Metric 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 section for details on each metric). Format: <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] Supported metrics Examples: Calculate the value of CrossEntropy CrossEntropy Calculate the value of Quantile with the coefficient α = 0.1 \alpha = 0.1 Quantile:alpha=0.1 Calculate the values of Logloss and AUC [ 'Logloss' , 'AUC' ] Values of all custom metrics for learn and validation datasets are saved to the Metric output files ( learn_error.tsv and test_error.tsv respectively). The directory for these files is specified in the --train-dir ( train_dir ) parameter. Use the visualization tools to see a live chart with the dynamics of the specified metrics. 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 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 A user-defined function can also be set as the value (see an example ). Examples: R2 The following parameters can be set for the corresponding methods and are used when the model is trained or applied. Parameters for trained or applied model The following parameters can be set for the corresponding methods and are used when the model is trained or applied. Classes: fit ( CatBoost ) fit ( CatBoostClassifier ) fit ( CatBoostRegressor ) 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. verbose Output the measured evaluation metric to stderr. plot Plot the following information during training: the metric values; the custom loss values; the loss function change during feature selection; the time has passed since training started; the remaining time until the end of training. This option can be used if training is performed in Jupyter notebook. R package The following parameters can be set for the corresponding methods and are used when the model is trained or applied. Method: catboost.train loss_function Description The metric 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 section for details on each metric). Format: <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] Supported metrics RMSE Logloss MAE CrossEntropy Quantile LogLinQuantile Lq MultiRMSE MultiClass MultiClassOneVsAll MultiLogloss MultiCrossEntropy MAPE Poisson PairLogit PairLogitPairwise QueryRMSE QuerySoftMax GroupQuantile Tweedie YetiRank YetiRankPairwise StochasticFilter StochasticRank For example, use the following construction to calculate the value of Quantile with the coefficient α = 0.1 \alpha = 0.1 : Quantile:alpha=0.1 custom_loss Parameters Metric 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 section for details on each metric). Format: <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] Supported metrics Examples: Calculate the value of CrossEntropy c('CrossEntropy') Or simply: 'CrossEntropy' Calculate the values of Logloss and AUC c('Logloss', 'AUC') Calculate the value of Quantile with the coefficient α = 0.1 \alpha = 0.1 c('Quantilealpha=0.1') Values of all custom metrics for learn and validation datasets are saved to the Metric output files ( learn_error.tsv and test_error.tsv respectively). The directory for these files is specified in the --train-dir ( train_dir ) parameter. 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 Parameters 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 Quantile:alpha=0.3 Command-line version The following command keys can be specified for the corresponding commands and are used when the model is trained or applied. Params for the catboost fit command: --loss-function The metric 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 section for details on each metric). Format: <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] Supported metrics RMSE Logloss MAE CrossEntropy Quantile LogLinQuantile Lq MultiRMSE MultiClass MultiClassOneVsAll MultiLogloss MultiCrossEntropy MAPE Poisson PairLogit PairLogitPairwise QueryRMSE QuerySoftMax GroupQuantile Tweedie YetiRank YetiRankPairwise StochasticFilter StochasticRank For example, use the following construction to calculate the value of Quantile with the coefficient α = 0.1 \alpha = 0.1 : Quantilealpha=0.1 --custom-metric Metric 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 section for details on each metric). Format: <Metric 1>[:<parameter 1>=<value>;..;<parameter N>=<value>],<Metric 2>[:<parameter 1>=<value>;..;<parameter N>=<value>],..,<Metric N>[:<parameter 1>=<value>;..;<parameter N>=<value>] Supported metrics Examples: Calculate the value of CrossEntropy CrossEntropy Calculate the value of Quantile with the coefficient α = 0.1 \alpha = 0.1 Quantilealpha=0.1 Values of all custom metrics for learn and validation datasets are saved to the Metric output files ( learn_error.tsv and test_error.tsv respectively). The directory for these files is specified in the --train-dir ( train_dir ) parameter. --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 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 Examples: R2 Quantile:alpha=0.3 --logging-level The logging level to output to stdout. Possible values: Silent — Do not output any logging information to stdout. Verbose — Output the following data to stdout: optimized metric elapsed time of training remaining time of training Info — Output additional information and the number of trees. Debug — Output debugging information.
Markdown	[![Logo icon](https://yastatic.net/s3/locdoc/daas-static/catboost/71b237a322eec6f2889af0dae2a9c549.svg)](https://catboost.ai/ "CatBoost") - Installation - [Overview](https://catboost.ai/docs/en/features/en/concepts/installation) - Python package installation - CatBoost for Apache Spark installation - R package installation - Command-line version binary - Build from source - Key Features - [Training](https://catboost.ai/docs/en/features/en/features/training) - [Training on GPU](https://catboost.ai/docs/en/features/en/features/training-on-gpu) - [Regular prediction](https://catboost.ai/docs/en/features/en/features/prediction) - [Staged prediction](https://catboost.ai/docs/en/features/en/features/staged-prediction) - [Cross-validation](https://catboost.ai/docs/en/features/en/features/cross-validation) - [Feature importances](https://catboost.ai/docs/en/features/en/features/feature-importances-calculation) - [User-defined metrics](https://catboost.ai/docs/en/features/en/features/custom-loss-functions) - [Using the overfitting detector](https://catboost.ai/docs/en/features/en/features/overfitting-detector-desc) - [Export a model to CoreML](https://catboost.ai/docs/en/features/en/features/export-model-to-core-ml) - [Pre-trained data](https://catboost.ai/docs/en/features/en/features/proceed-training) - [Calculate metrics](https://catboost.ai/docs/en/features/en/features/eval-metrics) - [Categorical features](https://catboost.ai/docs/en/features/en/features/categorical-features) - [Text features](https://catboost.ai/docs/en/features/en/features/text-features) - [Embeddings features](https://catboost.ai/docs/en/features/en/features/embeddings-features) - [Aggregated graph features](https://catboost.ai/docs/en/features/en/features/graph-aggregated-features) - [Implemented metrics](https://catboost.ai/docs/en/features/en/features/loss-functions-desc) - [Export a model to Python or C++](https://catboost.ai/docs/en/features/en/features/export-model-to-python) - [Export a model to JSON](https://catboost.ai/docs/en/features/en/features/export-model-to-json) - [Object importances](https://catboost.ai/docs/en/features/en/features/object-importances-calcution) - Training parameters - Python package - CatBoost for Apache Spark - R package - Command-line version - Applying models - Objectives and metrics - Model analysis - Data format description - [Parameter tuning](https://catboost.ai/docs/en/features/en/concepts/parameter-tuning) - [Speeding up the training](https://catboost.ai/docs/en/features/en/concepts/speed-up-training) - Data visualization - Algorithm details - [FAQ](https://catboost.ai/docs/en/features/en/concepts/faq) - Educational materials - [Development and contributions](https://catboost.ai/docs/en/features/en/concepts/development-and-contributions) - [Contacts](https://catboost.ai/docs/en/features/en/concepts/contacts) Implemented metrics ## In this article: - [Python package](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#python-package) - [Parameters for trained model](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#parameters-for-trained-model) - [Parameters for trained or applied model](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#parameters-for-trained-or-applied-model) - [R package](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#r-package) - [loss\_function](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#loss_function) - [custom\_loss](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#custom_loss) - [use-best-model](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#use-best-model1) - [eval-metric](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#eval-metric1) - [Command-line version](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#command-line-version) - [\--loss-function](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#loss-function1) - [\--custom-metric](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#custom-metric) - [\--use-best-model](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#use-best-model2) - [\--eval-metric](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#eval-metric2) - [\--logging-level](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#logging-level) 1. Key Features 2. Implemented metrics # Implemented metrics - [Python package](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#python-package) - [Parameters for trained model](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#parameters-for-trained-model) - [Parameters for trained or applied model](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#parameters-for-trained-or-applied-model) - [R package](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#r-package) - [loss\_function](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#loss_function) - [custom\_loss](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#custom_loss) - [use-best-model](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#use-best-model1) - [eval-metric](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#eval-metric1) - [Command-line version](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#command-line-version) - [\--loss-function](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#loss-function1) - [\--custom-metric](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#custom-metric) - [\--use-best-model](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#use-best-model2) - [\--eval-metric](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#eval-metric2) - [\--logging-level](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#logging-level) CatBoost provides built-in metrics for various machine learning problems. These functions can be used for model optimization or reference purposes. See the [Objectives and metrics](https://catboost.ai/docs/en/features/en/concepts/loss-functions) section for details on the calculation principles. Choose the implementation for more details. - [python](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#python) - [r-package](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#r-package) - [cli](https://catboost.ai/docs/en/features/en/features/loss-functions-desc#command-line-version) ## Python package The following parameters can be set for the corresponding classes and are used when the model is trained. ### Parameters for trained model Classes: - [CatBoost](https://catboost.ai/docs/en/features/en/concepts/python-reference_catboost) - [CatBoostClassifier](https://catboost.ai/docs/en/features/en/concepts/python-reference_catboostclassifier) - [CatBoostRegressor](https://catboost.ai/docs/en/features/en/concepts/python-reference_catboostregressor) #### loss-function The [metric](https://catboost.ai/docs/en/features/en/concepts/loss-functions) 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/features/en/concepts/loss-functions) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` Supported metrics - RMSE - Logloss - MAE - CrossEntropy - Quantile - LogLinQuantile - Lq - MultiRMSE - MultiClass - MultiClassOneVsAll - MultiLogloss - MultiCrossEntropy - MAPE - Poisson - PairLogit - PairLogitPairwise - QueryRMSE - QuerySoftMax - GroupQuantile - Tweedie - YetiRank - YetiRankPairwise - StochasticFilter - StochasticRank A custom python object can also be set as the value of this parameter (see an [example](https://catboost.ai/docs/en/features/en/concepts/python-usages-examples)). For example, use the following construction to calculate the value of Quantile with the coefficient α \= 0\.1 \\alpha = 0.1 α\=0\.1: ``` Quantile:alpha=0.1 ``` #### custom\_metric [Metric](https://catboost.ai/docs/en/features/en/concepts/loss-functions) 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/features/en/concepts/loss-functions) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` [Supported metrics](https://catboost.ai/docs/en/features/en/references/custom-metric__supported-metrics) Examples: - Calculate the value of CrossEntropy ``` CrossEntropy ``` - Calculate the value of Quantile with the coefficient α \= 0\.1 \\alpha = 0.1 α\=0\.1 ``` Quantile:alpha=0.1 ``` - Calculate the values of Logloss and AUC ``` ['Logloss', 'AUC'] ``` Values of all custom metrics for learn and validation datasets are saved to the [Metric](https://catboost.ai/docs/en/features/en/concepts/output-data_loss-function) output files (`learn_error.tsv` and `test_error.tsv` respectively). The directory for these files is specified in the `--train-dir` (`train_dir`) parameter. Use the [visualization tools](https://catboost.ai/docs/en/features/en/features/visualization) to see a live chart with the dynamics of the specified metrics. #### use-best-model If this parameter is set, the number of trees that are saved in the resulting model is defined as follows: 1. Build the number of trees defined by the training parameters. 2. 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 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/features/en/concepts/loss-functions) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` [Supported metrics](https://catboost.ai/docs/en/features/en/references/eval-metric__supported-metrics) A user-defined function can also be set as the value (see an [example](https://catboost.ai/docs/en/features/en/concepts/python-usages-examples)). Examples: ``` R2 ``` The following parameters can be set for the corresponding methods and are used when the model is trained or applied. ### Parameters for trained or applied model The following parameters can be set for the corresponding methods and are used when the model is trained or applied. Classes: - [fit](https://catboost.ai/docs/en/features/en/concepts/python-reference_catboost_fit) ([CatBoost](https://catboost.ai/docs/en/features/en/concepts/python-reference_catboost)) - [fit](https://catboost.ai/docs/en/features/en/concepts/python-reference_catboostclassifier_fit) ([CatBoostClassifier](https://catboost.ai/docs/en/features/en/concepts/python-reference_catboostclassifier)) - [fit](https://catboost.ai/docs/en/features/en/concepts/python-reference_catboostregressor_fit) ([CatBoostRegressor](https://catboost.ai/docs/en/features/en/concepts/python-reference_catboostregressor)) #### use\_best\_model If this parameter is set, the number of trees that are saved in the resulting model is defined as follows: 1. Build the number of trees defined by the training parameters. 2. 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. #### verbose Output the measured evaluation metric to stderr. #### plot Plot the following information during training: - the metric values; - the custom loss values; - the loss function change during feature selection; - the time has passed since training started; - the remaining time until the end of training. This [option can be used](https://catboost.ai/docs/en/features/en/features/visualization_jupyter-notebook) if training is performed in Jupyter notebook. ## R package The following parameters can be set for the corresponding methods and are used when the model is trained or applied. Method: [catboost.train](https://catboost.ai/docs/en/features/en/concepts/r-reference_catboost-train) ### loss\_function Description The [metric](https://catboost.ai/docs/en/features/en/concepts/loss-functions) 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/features/en/concepts/loss-functions) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` Supported metrics - RMSE - Logloss - MAE - CrossEntropy - Quantile - LogLinQuantile - Lq - MultiRMSE - MultiClass - MultiClassOneVsAll - MultiLogloss - MultiCrossEntropy - MAPE - Poisson - PairLogit - PairLogitPairwise - QueryRMSE - QuerySoftMax - GroupQuantile - Tweedie - YetiRank - YetiRankPairwise - StochasticFilter - StochasticRank For example, use the following construction to calculate the value of Quantile with the coefficient α \= 0\.1 \\alpha = 0.1 α\=0\.1: ``` Quantile:alpha=0.1 ``` ### custom\_loss Parameters [Metric](https://catboost.ai/docs/en/features/en/concepts/loss-functions) 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/features/en/concepts/loss-functions) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` [Supported metrics](https://catboost.ai/docs/en/features/en/references/custom-metric__supported-metrics) Examples: - Calculate the value of CrossEntropy ``` c('CrossEntropy') ``` Or simply: ``` 'CrossEntropy' ``` - Calculate the values of Logloss and AUC ``` c('Logloss', 'AUC') ``` - Calculate the value of Quantile with the coefficient α \= 0\.1 \\alpha = 0.1 α\=0\.1 ``` c('Quantilealpha=0.1') ``` Values of all custom metrics for learn and validation datasets are saved to the [Metric](https://catboost.ai/docs/en/features/en/concepts/output-data_loss-function) output files (`learn_error.tsv` and `test_error.tsv` respectively). The directory for these files is specified in the `--train-dir` (`train_dir`) parameter. ### use-best-model If this parameter is set, the number of trees that are saved in the resulting model is defined as follows: 1. Build the number of trees defined by the training parameters. 2. 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 Parameters 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/features/en/concepts/loss-functions) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` [Supported metrics](https://catboost.ai/docs/en/features/en/references/eval-metric__supported-metrics) ``` Quantile:alpha=0.3 ``` ## Command-line version The following command keys can be specified for the corresponding commands and are used when the model is trained or applied. Params for the [catboost fit](https://catboost.ai/docs/en/features/en/references/training-parameters/) command: ### \--loss-function The [metric](https://catboost.ai/docs/en/features/en/concepts/loss-functions) 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/features/en/concepts/loss-functions) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` Supported metrics - RMSE - Logloss - MAE - CrossEntropy - Quantile - LogLinQuantile - Lq - MultiRMSE - MultiClass - MultiClassOneVsAll - MultiLogloss - MultiCrossEntropy - MAPE - Poisson - PairLogit - PairLogitPairwise - QueryRMSE - QuerySoftMax - GroupQuantile - Tweedie - YetiRank - YetiRankPairwise - StochasticFilter - StochasticRank For example, use the following construction to calculate the value of Quantile with the coefficient α \= 0\.1 \\alpha = 0.1 α\=0\.1: ``` Quantilealpha=0.1 ``` ### \--custom-metric [Metric](https://catboost.ai/docs/en/features/en/concepts/loss-functions) 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/features/en/concepts/loss-functions) section for details on each metric). Format: ``` <Metric 1>[:<parameter 1>=<value>;..;<parameter N>=<value>],<Metric 2>[:<parameter 1>=<value>;..;<parameter N>=<value>],..,<Metric N>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` [Supported metrics](https://catboost.ai/docs/en/features/en/references/custom-metric__supported-metrics) Examples: - Calculate the value of CrossEntropy ``` CrossEntropy ``` - Calculate the value of Quantile with the coefficient α \= 0\.1 \\alpha = 0.1 α\=0\.1 ``` Quantilealpha=0.1 ``` Values of all custom metrics for learn and validation datasets are saved to the [Metric](https://catboost.ai/docs/en/features/en/concepts/output-data_loss-function) output files (`learn_error.tsv` and `test_error.tsv` respectively). The directory for these files is specified in the `--train-dir` (`train_dir`) parameter. ### \--use-best-model If this parameter is set, the number of trees that are saved in the resulting model is defined as follows: 1. Build the number of trees defined by the training parameters. 2. 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 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/features/en/concepts/loss-functions) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` [Supported metrics](https://catboost.ai/docs/en/features/en/references/eval-metric__supported-metrics) Examples: ``` R2 ``` ``` Quantile:alpha=0.3 ``` ### \--logging-level The logging level to output to stdout. Possible values: - Silent — Do not output any logging information to stdout. - Verbose — Output the following data to stdout: - optimized metric - elapsed time of training - remaining time of training - Info — Output additional information and the number of trees. - Debug — Output debugging information. ### Was the article helpful? Yes No Previous [Aggregated graph features](https://catboost.ai/docs/en/features/en/features/graph-aggregated-features) Next [Export a model to Python or C++](https://catboost.ai/docs/en/features/en/features/export-model-to-python) ![](https://mc.yandex.ru/watch/60763294)
Readable Markdown	CatBoost provides built-in metrics for various machine learning problems. These functions can be used for model optimization or reference purposes. See the [Objectives and metrics](https://catboost.ai/docs/en/concepts/loss-functions) section for details on the calculation principles. Choose the implementation for more details. - [python](https://catboost.ai/docs/en/features/loss-functions-desc#python) - [r-package](https://catboost.ai/docs/en/features/loss-functions-desc#r-package) - [cli](https://catboost.ai/docs/en/features/loss-functions-desc#command-line-version) ## Python package The following parameters can be set for the corresponding classes and are used when the model is trained. ### Parameters for trained model Classes: - [CatBoost](https://catboost.ai/docs/en/concepts/python-reference_catboost) - [CatBoostClassifier](https://catboost.ai/docs/en/concepts/python-reference_catboostclassifier) - [CatBoostRegressor](https://catboost.ai/docs/en/concepts/python-reference_catboostregressor) #### loss-function The [metric](https://catboost.ai/docs/en/concepts/loss-functions) 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) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` Supported metrics - RMSE - Logloss - MAE - CrossEntropy - Quantile - LogLinQuantile - Lq - MultiRMSE - MultiClass - MultiClassOneVsAll - MultiLogloss - MultiCrossEntropy - MAPE - Poisson - PairLogit - PairLogitPairwise - QueryRMSE - QuerySoftMax - GroupQuantile - Tweedie - YetiRank - YetiRankPairwise - StochasticFilter - StochasticRank A custom python object can also be set as the value of this parameter (see an [example](https://catboost.ai/docs/en/concepts/python-usages-examples)). For example, use the following construction to calculate the value of Quantile with the coefficient α \= 0\.1 \\alpha = 0.1: ``` Quantile:alpha=0.1 ``` #### custom\_metric [Metric](https://catboost.ai/docs/en/concepts/loss-functions) 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) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` [Supported metrics](https://catboost.ai/docs/en/references/custom-metric__supported-metrics) Examples: - Calculate the value of CrossEntropy ``` CrossEntropy ``` - Calculate the value of Quantile with the coefficient α \= 0\.1 \\alpha = 0.1 ``` Quantile:alpha=0.1 ``` - Calculate the values of Logloss and AUC ``` ['Logloss', 'AUC'] ``` Values of all custom metrics for learn and validation datasets are saved to the [Metric](https://catboost.ai/docs/en/concepts/output-data_loss-function) output files (`learn_error.tsv` and `test_error.tsv` respectively). The directory for these files is specified in the `--train-dir` (`train_dir`) parameter. Use the [visualization tools](https://catboost.ai/docs/en/features/visualization) to see a live chart with the dynamics of the specified metrics. #### use-best-model If this parameter is set, the number of trees that are saved in the resulting model is defined as follows: 1. Build the number of trees defined by the training parameters. 2. 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 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) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` [Supported metrics](https://catboost.ai/docs/en/references/eval-metric__supported-metrics) A user-defined function can also be set as the value (see an [example](https://catboost.ai/docs/en/concepts/python-usages-examples)). Examples: ``` R2 ``` The following parameters can be set for the corresponding methods and are used when the model is trained or applied. ### Parameters for trained or applied model The following parameters can be set for the corresponding methods and are used when the model is trained or applied. Classes: - [fit](https://catboost.ai/docs/en/concepts/python-reference_catboost_fit) ([CatBoost](https://catboost.ai/docs/en/concepts/python-reference_catboost)) - [fit](https://catboost.ai/docs/en/concepts/python-reference_catboostclassifier_fit) ([CatBoostClassifier](https://catboost.ai/docs/en/concepts/python-reference_catboostclassifier)) - [fit](https://catboost.ai/docs/en/concepts/python-reference_catboostregressor_fit) ([CatBoostRegressor](https://catboost.ai/docs/en/concepts/python-reference_catboostregressor)) #### use\_best\_model If this parameter is set, the number of trees that are saved in the resulting model is defined as follows: 1. Build the number of trees defined by the training parameters. 2. 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. #### verbose Output the measured evaluation metric to stderr. #### plot Plot the following information during training: - the metric values; - the custom loss values; - the loss function change during feature selection; - the time has passed since training started; - the remaining time until the end of training. This [option can be used](https://catboost.ai/docs/en/features/visualization_jupyter-notebook) if training is performed in Jupyter notebook. ## R package The following parameters can be set for the corresponding methods and are used when the model is trained or applied. Method: [catboost.train](https://catboost.ai/docs/en/concepts/r-reference_catboost-train) ### loss\_function Description The [metric](https://catboost.ai/docs/en/concepts/loss-functions) 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) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` Supported metrics - RMSE - Logloss - MAE - CrossEntropy - Quantile - LogLinQuantile - Lq - MultiRMSE - MultiClass - MultiClassOneVsAll - MultiLogloss - MultiCrossEntropy - MAPE - Poisson - PairLogit - PairLogitPairwise - QueryRMSE - QuerySoftMax - GroupQuantile - Tweedie - YetiRank - YetiRankPairwise - StochasticFilter - StochasticRank For example, use the following construction to calculate the value of Quantile with the coefficient α \= 0\.1 \\alpha = 0.1: ``` Quantile:alpha=0.1 ``` ### custom\_loss Parameters [Metric](https://catboost.ai/docs/en/concepts/loss-functions) 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) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` [Supported metrics](https://catboost.ai/docs/en/references/custom-metric__supported-metrics) Examples: - Calculate the value of CrossEntropy ``` c('CrossEntropy') ``` Or simply: ``` 'CrossEntropy' ``` - Calculate the values of Logloss and AUC ``` c('Logloss', 'AUC') ``` - Calculate the value of Quantile with the coefficient α \= 0\.1 \\alpha = 0.1 ``` c('Quantilealpha=0.1') ``` Values of all custom metrics for learn and validation datasets are saved to the [Metric](https://catboost.ai/docs/en/concepts/output-data_loss-function) output files (`learn_error.tsv` and `test_error.tsv` respectively). The directory for these files is specified in the `--train-dir` (`train_dir`) parameter. ### use-best-model If this parameter is set, the number of trees that are saved in the resulting model is defined as follows: 1. Build the number of trees defined by the training parameters. 2. 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 Parameters 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) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` [Supported metrics](https://catboost.ai/docs/en/references/eval-metric__supported-metrics) ``` Quantile:alpha=0.3 ``` ## Command-line version The following command keys can be specified for the corresponding commands and are used when the model is trained or applied. Params for the [catboost fit](https://catboost.ai/docs/en/references/training-parameters/) command: ### \--loss-function The [metric](https://catboost.ai/docs/en/concepts/loss-functions) 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) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` Supported metrics - RMSE - Logloss - MAE - CrossEntropy - Quantile - LogLinQuantile - Lq - MultiRMSE - MultiClass - MultiClassOneVsAll - MultiLogloss - MultiCrossEntropy - MAPE - Poisson - PairLogit - PairLogitPairwise - QueryRMSE - QuerySoftMax - GroupQuantile - Tweedie - YetiRank - YetiRankPairwise - StochasticFilter - StochasticRank For example, use the following construction to calculate the value of Quantile with the coefficient α \= 0\.1 \\alpha = 0.1: ``` Quantilealpha=0.1 ``` ### \--custom-metric [Metric](https://catboost.ai/docs/en/concepts/loss-functions) 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) section for details on each metric). Format: ``` <Metric 1>[:<parameter 1>=<value>;..;<parameter N>=<value>],<Metric 2>[:<parameter 1>=<value>;..;<parameter N>=<value>],..,<Metric N>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` [Supported metrics](https://catboost.ai/docs/en/references/custom-metric__supported-metrics) Examples: - Calculate the value of CrossEntropy ``` CrossEntropy ``` - Calculate the value of Quantile with the coefficient α \= 0\.1 \\alpha = 0.1 ``` Quantilealpha=0.1 ``` Values of all custom metrics for learn and validation datasets are saved to the [Metric](https://catboost.ai/docs/en/concepts/output-data_loss-function) output files (`learn_error.tsv` and `test_error.tsv` respectively). The directory for these files is specified in the `--train-dir` (`train_dir`) parameter. ### \--use-best-model If this parameter is set, the number of trees that are saved in the resulting model is defined as follows: 1. Build the number of trees defined by the training parameters. 2. 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 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) section for details on each metric). Format: ``` <Metric>[:<parameter 1>=<value>;..;<parameter N>=<value>] ``` [Supported metrics](https://catboost.ai/docs/en/references/eval-metric__supported-metrics) Examples: ``` R2 ``` ``` Quantile:alpha=0.3 ``` ### \--logging-level The logging level to output to stdout. Possible values: - Silent — Do not output any logging information to stdout. - Verbose — Output the following data to stdout: - optimized metric - elapsed time of training - remaining time of training - Info — Output additional information and the number of trees. - Debug — Output debugging information.
Shard	169 (laksa)
Root Hash	17435841955170310369
Unparsed URL	ai,catboost!/docs/en/features/loss-functions-desc s443