Expert SettingsΒΆ
This section describes the Expert Settings that are available when starting an experiment. Driverless AI provides a variety of options in the Expert Settings that let you customize your experiment. Use the search bar to refine the list of settings or locate a specific setting.
The default values for these options are derived from the configuration options in the config.toml file. For more information about each of these options, see the Sample config.toml File section. When a setting is changed from the default value, it is highlighted in the interface to indicate that the default value is not currently selected.
Note about Feature Brain Level: By default, the feature brain pulls in any better model regardless of the features even if the new model disabled those features. For full control over features pulled in via changes in these Expert Settings, users should set the Feature Brain Level option to 0.
- Upload Custom Recipe
- Load Custom Recipe from URL
- Official Recipes (Open Source)
- Edit the TOML Configuration
- Experiment Settings
max_runtime_minutesmax_runtime_minutes_until_abortpipeline-building-recipeenable_genetic_algorithmtournament_stylemake_python_scoring_pipelinemake_mojo_scoring_pipelinereduce_mojo_sizebenchmark_mojo_latencymojo_building_timeoutmojo_building_parallelismmake_pipeline_visualizationmake_autoreportmin_num_rowskaggle_usernamekaggle_keykaggle_timeoutreproducibility_levelseedallow_different_classes_across_fold_splitsmax_num_classesmax_num_classes_compute_rocmax_num_classes_client_and_guiroc_reduce_typefeature_brain1feature_brain2feature_brain3feature_brain4feature_brain5force_model_restart_to_defaultsmin_dai_iterationstarget_transformerfixed_num_folds_evolutionfixed_num_foldsfixed_only_first_fold_modelfeature_evolution_data_sizefinal_pipeline_data_sizemax_validation_to_training_size_ratio_for_final_ensembleforce_stratified_splits_for_imbalanced_threshold_binarymli_customlast_recipetime_abort
- Model Settings
enable_constant_modelenable_decision_treeenable_glmenable_xgboost_gbmenable_lightgbmenable_xgboost_dartenable_xgboost_rapidsenable_xgboost_rfenable_xgboost_gbm_daskenable_xgboost_dart_daskenable_lightgbm_daskenable_hyperopt_dasknum_inner_hyperopt_trials_prefinalnum_inner_hyperopt_trials_finalnum_hyperopt_individuals_finaloptuna_pruneroptuna_samplerenable_xgboost_hyperopt_callbackenable_lightgbm_hyperopt_callbackenable_tensorflowenable_pytorchenable_ftrlenable_rulefitenable_zero_inflated_modelsenable_lightgbm_boosting_typesenable_lightgbm_cuda_supportshow_constant_modelparams_tensorflowmax_nestimatorsn_estimators_list_no_early_stoppingmin_learning_rate_finalmax_learning_rate_finalmax_nestimators_feature_evolution_factormin_learning_ratemax_learning_ratemax_epochsmax_max_depthmax_max_binrulefit_max_num_rulesfixed_ensemble_levelcross_validate_single_final_modelparameter_tuning_num_modelsimbalance_sampling_methodimbalance_sampling_threshold_min_rows_originalimbalance_ratio_sampling_thresholdheavy_imbalance_ratio_sampling_thresholdimbalance_sampling_number_of_bagsimbalance_sampling_max_number_of_bagsimbalance_sampling_max_number_of_bags_feature_evolutionimbalance_sampling_max_multiple_data_sizeimbalance_sampling_target_minority_fractionftrl_max_interaction_terms_per_degreeenable_bootstraptensorflow_num_classes_switchprediction_intervalsprediction_intervals_alpha
- Features Settings
feature_engineering_effortcheck_distribution_shiftcheck_distribution_shift_dropdrop_features_distribution_shift_threshold_auccheck_leakagedrop_features_leakage_threshold_aucleakage_max_data_sizeenable_wide_rulesorig_features_fs_reportmax_rows_fsmax_orig_cols_selectedmax_orig_nonnumeric_cols_selectedfs_orig_cols_selectedfs_orig_numeric_cols_selectedfs_orig_nonnumeric_cols_selectedmax_relative_cardinalitynum_as_catnfeatures_maxngenes_maxfeatures_allowed_by_interpretabilitymonotonicity_constraints_interpretability_switchmonotonicity_constraints_correlation_thresholdmonotonicity_constraints_log_levelmonotonicity_constraints_drop_low_correlation_featuresmonotonicity_constraints_dictmax_feature_interaction_depthfixed_feature_interaction_depthenable_target_encodingcvte_cv_in_cvenable_lexilabel_encodingenable_isolation_forestenable_one_hot_encodingisolation_forest_nestimatorsdrop_constant_columnsdrop_id_columnsno_drop_featurescols_to_dropcols_to_group_bysample_cols_to_group_byagg_funcs_for_group_byfolds_for_group_bymutation_modedump_varimp_every_scored_indivdump_trans_timingscompute_correlationinteraction_finder_gini_rel_improvement_thresholdinteraction_finder_return_limitenable_rapids_transformers
- Time Series Settings
time_series_recipetime_series_merge_splitsmerge_splits_max_valid_ratiofixed_size_splitstime_series_validation_fold_split_datetime_boundariestimeseries_split_suggestion_timeoutholiday_featuresholiday_countriesoverride_lag_sizesoverride_ufapt_lag_sizesoverride_non_ufapt_lag_sizesmin_lag_sizeallow_time_column_as_featureallow_time_column_as_numeric_featuredatetime_funcsfilter_datetime_funcsallow_tgc_as_featuresallowed_coltypes_for_tgc_as_featuresenable_time_unaware_transformerstgc_only_use_all_groupstime_series_holdout_predstime_series_validation_splitstime_series_splits_max_overlaptime_series_max_holdout_splitsmli_ts_fast_approxmli_ts_fast_approx_contribsmli_ts_holdout_contribstime_series_min_interpretabilitylags_dropoutprob_lag_non_targetsrolling_test_methodfast_tta_internalprob_default_lagsprob_lagsinteractionprob_lagsaggregatests_target_trafots_target_trafo_epidemic_params_dictts_target_trafo_epidemic_targetts_lag_target_trafots_target_trafo_lag_size
- NLP Settings
tensorflow_max_epochs_nlpenable_tensorflow_nlp_accuracy_switchenable_tensorflow_textcnnenable_tensorflow_textbigruenable_tensorflow_charcnnenable_pytorch_nlppytorch_nlp_pretrained_modelspytorch_nlp_fine_tuning_num_epochspytorch_nlp_fine_tuning_batch_sizepytorch_nlp_fine_tuning_padding_lengthpytorch_nlp_pretrained_models_dirtensorflow_nlp_pretrained_embeddings_file_pathtensorflow_nlp_pretrained_embeddings_trainabletext_fraction_for_text_dominated_problemtext_transformer_fraction_for_text_dominated_problemstring_col_as_text_thresholdtext_transformers_max_vocabulary_size
- Image Settings
enable_tensorflow_imagetensorflow_image_pretrained_modelstensorflow_image_vectorization_output_dimensiontensorflow_image_fine_tunetensorflow_image_fine_tuning_num_epochstensorflow_image_augmentationstensorflow_image_batch_sizeimage_download_timeoutstring_col_as_image_max_missing_fractionstring_col_as_image_min_valid_types_fractiontensorflow_image_use_gpu
- Recipes Settings
included_transformersincluded_pretransformersnum_pipeline_layersincluded_modelsincluded_scorersthreshold_scorerincluded_datasprob_add_genesprob_addbest_genesprob_prune_genesprob_perturb_xgbprob_prune_by_featuresacceptance_test_timeoutskip_transformer_failuresskip_model_failuresdetailed_skip_failure_messages_level
- System Settings
exclusive_modemax_coresmax_fit_coresuse_dask_clustermax_predict_coresmax_predict_cores_in_daibatch_cpu_tuning_max_workerscpu_max_workersnum_gpus_per_experimentmin_num_cores_per_gpunum_gpus_per_modelnum_gpus_for_predictionassumed_simultaneous_dt_forks_mungingmax_max_dt_threads_mungingmax_dt_threads_mungingmax_dt_threads_readwritemax_dt_threads_stats_openblasgpu_id_startallow_reduce_features_when_failurereduce_repeats_when_failurefraction_anchor_reduce_features_when_failuredetailed_tracesdebug_loglog_system_info_per_experiment
- AutoDoc Settings
make_autoreportautodoc_report_nameautodoc_templateautodoc_output_typeautodoc_subtemplate_typeautodoc_max_cm_sizeautodoc_num_featuresautodoc_min_relative_importanceautodoc_include_permutation_feature_importanceautodoc_feature_importance_num_permautodoc_feature_importance_scorerautodoc_pd_max_rowsautodoc_pd_max_runtimeautodoc_out_of_rangeautodoc_num_rowsautodoc_population_stability_indexautodoc_population_stability_index_n_quantilesautodoc_prediction_statsautodoc_prediction_stats_n_quantilesautodoc_response_rateautodoc_response_rate_n_quantilesautodoc_gini_plotautodoc_enable_shapley_valuesautodoc_data_summary_col_numautodoc_list_all_config_settingsautodoc_keras_summary_line_lengthautodoc_transformer_architecture_max_linesautodoc_full_architecture_in_appendixautodoc_coef_table_appendix_results_tableautodoc_coef_table_num_modelsautodoc_coef_table_num_foldsautodoc_coef_table_num_coefautodoc_coef_table_num_classesautodoc_num_histogram_plots