public interface Grid
| Modifier and Type | Interface and Description |
|---|---|
static class |
Grid.Helper |
| Modifier and Type | Method and Description |
|---|---|
retrofit2.Call<AggregatorV99> |
trainAggregator() |
retrofit2.Call<AggregatorV99> |
trainAggregator(DataInfoTransformType transform,
PCAMethod pca_method,
int k,
int max_iterations,
int target_num_exemplars,
double rel_tol_num_exemplars,
long seed,
boolean use_all_factor_levels,
boolean save_mapping_frame,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for Aggregator model.
|
retrofit2.Call<CoxPHV3> |
trainCoxph() |
retrofit2.Call<CoxPHV3> |
trainCoxph(java.lang.String rcall,
java.lang.String stop_column,
java.lang.String start_column,
CoxPHTies ties,
double init,
double lre_min,
int iter_max,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for CoxPH model.
|
retrofit2.Call<DeepLearningV3> |
trainDeeplearning() |
retrofit2.Call<DeepLearningV3> |
trainDeeplearning(boolean balance_classes,
float[] class_sampling_factors,
float max_after_balance_size,
int max_confusion_matrix_size,
int max_hit_ratio_k,
DeepLearningActivation activation,
int[] hidden,
double epochs,
long train_samples_per_iteration,
double target_ratio_comm_to_comp,
long seed,
boolean adaptive_rate,
double rho,
double epsilon,
double rate,
double rate_annealing,
double rate_decay,
double momentum_start,
double momentum_ramp,
double momentum_stable,
boolean nesterov_accelerated_gradient,
double input_dropout_ratio,
double[] hidden_dropout_ratios,
double l1,
double l2,
float max_w2,
DeepLearningInitialWeightDistribution initial_weight_distribution,
double initial_weight_scale,
java.lang.String[] initial_weights,
java.lang.String[] initial_biases,
DeepLearningLoss loss,
double score_interval,
long score_training_samples,
long score_validation_samples,
double score_duty_cycle,
double classification_stop,
double regression_stop,
boolean quiet_mode,
DeepLearningClassSamplingMethod score_validation_sampling,
boolean overwrite_with_best_model,
boolean autoencoder,
boolean use_all_factor_levels,
boolean standardize,
boolean diagnostics,
boolean variable_importances,
boolean fast_mode,
boolean force_load_balance,
boolean replicate_training_data,
boolean single_node_mode,
boolean shuffle_training_data,
DeepLearningMissingValuesHandling missing_values_handling,
boolean sparse,
boolean col_major,
double average_activation,
double sparsity_beta,
int max_categorical_features,
boolean reproducible,
boolean export_weights_and_biases,
int mini_batch_size,
boolean elastic_averaging,
double elastic_averaging_moving_rate,
double elastic_averaging_regularization,
java.lang.String pretrained_autoencoder,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for DeepLearning model.
|
retrofit2.Call<DeepWaterV3> |
trainDeepwater() |
retrofit2.Call<DeepWaterV3> |
trainDeepwater(DeepWaterParametersProblemType problem_type,
DeepWaterParametersActivation activation,
int[] hidden,
double input_dropout_ratio,
double[] hidden_dropout_ratios,
int max_confusion_matrix_size,
boolean sparse,
int max_hit_ratio_k,
double epochs,
long train_samples_per_iteration,
double target_ratio_comm_to_comp,
long seed,
double learning_rate,
double learning_rate_annealing,
double momentum_start,
double momentum_ramp,
double momentum_stable,
double score_interval,
long score_training_samples,
long score_validation_samples,
double score_duty_cycle,
double classification_stop,
double regression_stop,
boolean quiet_mode,
boolean overwrite_with_best_model,
boolean autoencoder,
boolean diagnostics,
boolean variable_importances,
boolean replicate_training_data,
boolean single_node_mode,
boolean shuffle_training_data,
int mini_batch_size,
double clip_gradient,
DeepWaterParametersNetwork network,
DeepWaterParametersBackend backend,
int[] image_shape,
int channels,
boolean gpu,
int[] device_id,
boolean cache_data,
java.lang.String network_definition_file,
java.lang.String network_parameters_file,
java.lang.String mean_image_file,
java.lang.String export_native_parameters_prefix,
boolean standardize,
boolean balance_classes,
float[] class_sampling_factors,
float max_after_balance_size,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for DeepWater model.
|
retrofit2.Call<DRFV3> |
trainDrf() |
retrofit2.Call<DRFV3> |
trainDrf(int mtries,
boolean binomial_double_trees,
boolean balance_classes,
float[] class_sampling_factors,
float max_after_balance_size,
int max_confusion_matrix_size,
int max_hit_ratio_k,
int ntrees,
int max_depth,
double min_rows,
int nbins,
int nbins_top_level,
int nbins_cats,
double r2_stopping,
long seed,
boolean build_tree_one_node,
double sample_rate,
double[] sample_rate_per_class,
double col_sample_rate_per_tree,
double col_sample_rate_change_per_level,
int score_tree_interval,
double min_split_improvement,
TreeSharedTreeModelSharedTreeParametersHistogramType histogram_type,
boolean calibrate_model,
java.lang.String calibration_frame,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for DRF model.
|
retrofit2.Call<GBMV3> |
trainGbm() |
retrofit2.Call<GBMV3> |
trainGbm(double learn_rate,
double learn_rate_annealing,
double col_sample_rate,
double max_abs_leafnode_pred,
double pred_noise_bandwidth,
boolean balance_classes,
float[] class_sampling_factors,
float max_after_balance_size,
int max_confusion_matrix_size,
int max_hit_ratio_k,
int ntrees,
int max_depth,
double min_rows,
int nbins,
int nbins_top_level,
int nbins_cats,
double r2_stopping,
long seed,
boolean build_tree_one_node,
double sample_rate,
double[] sample_rate_per_class,
double col_sample_rate_per_tree,
double col_sample_rate_change_per_level,
int score_tree_interval,
double min_split_improvement,
TreeSharedTreeModelSharedTreeParametersHistogramType histogram_type,
boolean calibrate_model,
java.lang.String calibration_frame,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for GBM model.
|
retrofit2.Call<GLMV3> |
trainGlm() |
retrofit2.Call<GLMV3> |
trainGlm(long seed,
GLMFamily family,
double tweedie_variance_power,
double tweedie_link_power,
GLMSolver solver,
double[] alpha,
double[] lambda,
boolean lambda_search,
boolean early_stopping,
int nlambdas,
boolean standardize,
DeepLearningMissingValuesHandling missing_values_handling,
boolean non_negative,
int max_iterations,
double beta_epsilon,
double objective_epsilon,
double gradient_epsilon,
double obj_reg,
GLMLink link,
boolean intercept,
double prior,
double lambda_min_ratio,
java.lang.String beta_constraints,
int max_active_predictors,
java.lang.String[] interactions,
StringPairV3[] interaction_pairs,
boolean balance_classes,
float[] class_sampling_factors,
float max_after_balance_size,
int max_confusion_matrix_size,
int max_hit_ratio_k,
boolean compute_p_values,
boolean remove_collinear_columns,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for GLM model.
|
retrofit2.Call<GLRMV3> |
trainGlrm(DataInfoTransformType transform,
int k,
GenmodelalgosglrmGlrmLoss loss,
GenmodelalgosglrmGlrmLoss multi_loss,
GenmodelalgosglrmGlrmLoss[] loss_by_col,
int[] loss_by_col_idx,
int period,
GenmodelalgosglrmGlrmRegularizer regularization_x,
GenmodelalgosglrmGlrmRegularizer regularization_y,
double gamma_x,
double gamma_y,
int max_iterations,
int max_updates,
double init_step_size,
double min_step_size,
long seed,
GenmodelalgosglrmGlrmInitialization init,
SVDMethod svd_method,
java.lang.String user_y,
java.lang.String user_x,
java.lang.String loading_name,
boolean expand_user_y,
boolean impute_original,
boolean recover_svd,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for GLRM model.
|
retrofit2.Call<GLRMV3> |
trainGlrm(int k) |
retrofit2.Call<KMeansV3> |
trainKmeans() |
retrofit2.Call<KMeansV3> |
trainKmeans(java.lang.String user_points,
int max_iterations,
boolean standardize,
long seed,
KMeansInitialization init,
boolean estimate_k,
int k,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for KMeans model.
|
retrofit2.Call<NaiveBayesV3> |
trainNaivebayes() |
retrofit2.Call<NaiveBayesV3> |
trainNaivebayes(boolean balance_classes,
float[] class_sampling_factors,
float max_after_balance_size,
int max_confusion_matrix_size,
int max_hit_ratio_k,
double laplace,
double min_sdev,
double eps_sdev,
double min_prob,
double eps_prob,
boolean compute_metrics,
long seed,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for NaiveBayes model.
|
retrofit2.Call<PCAV3> |
trainPca(DataInfoTransformType transform,
PCAMethod pca_method,
int k,
int max_iterations,
long seed,
boolean use_all_factor_levels,
boolean compute_metrics,
boolean impute_missing,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for PCA model.
|
retrofit2.Call<PCAV3> |
trainPca(int k) |
retrofit2.Call<StackedEnsembleV99> |
trainStackedensemble(java.lang.String[] base_models) |
retrofit2.Call<StackedEnsembleV99> |
trainStackedensemble(java.lang.String[] base_models,
StackedEnsembleModelStackedEnsembleParametersMetalearnerAlgorithm metalearner_algorithm,
int metalearner_nfolds,
ModelParametersFoldAssignmentScheme metalearner_fold_assignment,
java.lang.String metalearner_fold_column,
boolean keep_levelone_frame,
java.lang.String metalearner_params,
long seed,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for StackedEnsemble model.
|
retrofit2.Call<SVDV99> |
trainSvd() |
retrofit2.Call<SVDV99> |
trainSvd(DataInfoTransformType transform,
SVDMethod svd_method,
int nv,
int max_iterations,
long seed,
boolean keep_u,
java.lang.String u_name,
boolean use_all_factor_levels,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for SVD model.
|
retrofit2.Call<Word2VecV3> |
trainWord2vec() |
retrofit2.Call<Word2VecV3> |
trainWord2vec(int vec_size,
int window_size,
float sent_sample_rate,
Word2VecNormModel norm_model,
int epochs,
int min_word_freq,
float init_learning_rate,
Word2VecWordModel word_model,
java.lang.String pre_trained,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for Word2Vec model.
|
retrofit2.Call<XGBoostV3> |
trainXgboost() |
retrofit2.Call<XGBoostV3> |
trainXgboost(int ntrees,
int n_estimators,
int max_depth,
double min_rows,
double min_child_weight,
double learn_rate,
double eta,
double sample_rate,
double subsample,
double col_sample_rate,
double colsample_bylevel,
double col_sample_rate_per_tree,
double colsample_bytree,
float max_abs_leafnode_pred,
float max_delta_step,
int score_tree_interval,
long seed,
float min_split_improvement,
float gamma,
int max_bins,
int max_leaves,
float min_sum_hessian_in_leaf,
float min_data_in_leaf,
TreexgboostXGBoostModelXGBoostParametersTreeMethod tree_method,
TreexgboostXGBoostModelXGBoostParametersGrowPolicy grow_policy,
TreexgboostXGBoostModelXGBoostParametersBooster booster,
float reg_lambda,
float reg_alpha,
boolean quiet_mode,
TreexgboostXGBoostModelXGBoostParametersDartSampleType sample_type,
TreexgboostXGBoostModelXGBoostParametersDartNormalizeType normalize_type,
float rate_drop,
boolean one_drop,
float skip_drop,
TreexgboostXGBoostModelXGBoostParametersDMatrixType dmatrix_type,
TreexgboostXGBoostModelXGBoostParametersBackend backend,
int gpu_id,
java.lang.String model_id,
java.lang.String training_frame,
java.lang.String validation_frame,
int nfolds,
boolean keep_cross_validation_predictions,
boolean keep_cross_validation_fold_assignment,
boolean parallelize_cross_validation,
GenmodelutilsDistributionFamily distribution,
double tweedie_power,
double quantile_alpha,
double huber_alpha,
java.lang.String response_column,
java.lang.String weights_column,
java.lang.String offset_column,
java.lang.String fold_column,
ModelParametersFoldAssignmentScheme fold_assignment,
ModelParametersCategoricalEncodingScheme categorical_encoding,
int max_categorical_levels,
java.lang.String[] ignored_columns,
boolean ignore_const_cols,
boolean score_each_iteration,
java.lang.String checkpoint,
int stopping_rounds,
double max_runtime_secs,
ScoreKeeperStoppingMetric stopping_metric,
double stopping_tolerance,
java.lang.String custom_metric_func)
Run grid search for XGBoost model.
|
@FormUrlEncoded @POST(value="/99/Grid/xgboost") retrofit2.Call<XGBoostV3> trainXgboost(@Field(value="ntrees") int ntrees, @Field(value="n_estimators") int n_estimators, @Field(value="max_depth") int max_depth, @Field(value="min_rows") double min_rows, @Field(value="min_child_weight") double min_child_weight, @Field(value="learn_rate") double learn_rate, @Field(value="eta") double eta, @Field(value="sample_rate") double sample_rate, @Field(value="subsample") double subsample, @Field(value="col_sample_rate") double col_sample_rate, @Field(value="colsample_bylevel") double colsample_bylevel, @Field(value="col_sample_rate_per_tree") double col_sample_rate_per_tree, @Field(value="colsample_bytree") double colsample_bytree, @Field(value="max_abs_leafnode_pred") float max_abs_leafnode_pred, @Field(value="max_delta_step") float max_delta_step, @Field(value="score_tree_interval") int score_tree_interval, @Field(value="seed") long seed, @Field(value="min_split_improvement") float min_split_improvement, @Field(value="gamma") float gamma, @Field(value="max_bins") int max_bins, @Field(value="max_leaves") int max_leaves, @Field(value="min_sum_hessian_in_leaf") float min_sum_hessian_in_leaf, @Field(value="min_data_in_leaf") float min_data_in_leaf, @Field(value="tree_method") TreexgboostXGBoostModelXGBoostParametersTreeMethod tree_method, @Field(value="grow_policy") TreexgboostXGBoostModelXGBoostParametersGrowPolicy grow_policy, @Field(value="booster") TreexgboostXGBoostModelXGBoostParametersBooster booster, @Field(value="reg_lambda") float reg_lambda, @Field(value="reg_alpha") float reg_alpha, @Field(value="quiet_mode") boolean quiet_mode, @Field(value="sample_type") TreexgboostXGBoostModelXGBoostParametersDartSampleType sample_type, @Field(value="normalize_type") TreexgboostXGBoostModelXGBoostParametersDartNormalizeType normalize_type, @Field(value="rate_drop") float rate_drop, @Field(value="one_drop") boolean one_drop, @Field(value="skip_drop") float skip_drop, @Field(value="dmatrix_type") TreexgboostXGBoostModelXGBoostParametersDMatrixType dmatrix_type, @Field(value="backend") TreexgboostXGBoostModelXGBoostParametersBackend backend, @Field(value="gpu_id") int gpu_id, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
ntrees - (same as n_estimators) Number of trees.n_estimators - (same as ntrees) Number of trees.max_depth - Maximum tree depth.min_rows - (same as min_child_weight) Fewest allowed (weighted) observations in a leaf.min_child_weight - (same as min_rows) Fewest allowed (weighted) observations in a leaf.learn_rate - (same as eta) Learning rate (from 0.0 to 1.0)eta - (same as learn_rate) Learning rate (from 0.0 to 1.0)sample_rate - (same as subsample) Row sample rate per tree (from 0.0 to 1.0)subsample - (same as sample_rate) Row sample rate per tree (from 0.0 to 1.0)col_sample_rate - (same as colsample_bylevel) Column sample rate (from 0.0 to 1.0)colsample_bylevel - (same as col_sample_rate) Column sample rate (from 0.0 to 1.0)col_sample_rate_per_tree - (same as colsample_bytree) Column sample rate per tree (from 0.0 to 1.0)colsample_bytree - (same as col_sample_rate_per_tree) Column sample rate per tree (from 0.0 to 1.0)max_abs_leafnode_pred - (same as max_delta_step) Maximum absolute value of a leaf node predictionmax_delta_step - (same as max_abs_leafnode_pred) Maximum absolute value of a leaf node predictionscore_tree_interval - Score the model after every so many trees. Disabled if set to 0.seed - Seed for pseudo random number generator (if applicable)min_split_improvement - (same as gamma) Minimum relative improvement in squared error reduction for a split
to happengamma - (same as min_split_improvement) Minimum relative improvement in squared error reduction for a split
to happenmax_bins - For tree_method=hist only: maximum number of binsmax_leaves - For tree_method=hist only: maximum number of leavesmin_sum_hessian_in_leaf - For tree_method=hist only: the mininum sum of hessian in a leaf to keep splittingmin_data_in_leaf - For tree_method=hist only: the mininum data in a leaf to keep splittingtree_method - Tree methodgrow_policy - Grow policy - depthwise is standard GBM, lossguide is LightGBMbooster - Booster typereg_lambda - L2 regularizationreg_alpha - L1 regularizationquiet_mode - Enable quiet modesample_type - For booster=dart only: sample_typenormalize_type - For booster=dart only: normalize_typerate_drop - For booster=dart only: rate_drop (0..1)one_drop - For booster=dart only: one_dropskip_drop - For booster=dart only: skip_drop (0..1)dmatrix_type - Type of DMatrix. For sparse, NAs and 0 are treated equally.backend - Backend. By default (auto), a GPU is used if available.gpu_id - Which GPU to use.model_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/xgboost") retrofit2.Call<XGBoostV3> trainXgboost()
@FormUrlEncoded @POST(value="/99/Grid/deeplearning") retrofit2.Call<DeepLearningV3> trainDeeplearning(@Field(value="balance_classes") boolean balance_classes, @Field(value="class_sampling_factors") float[] class_sampling_factors, @Field(value="max_after_balance_size") float max_after_balance_size, @Field(value="max_confusion_matrix_size") int max_confusion_matrix_size, @Field(value="max_hit_ratio_k") int max_hit_ratio_k, @Field(value="activation") DeepLearningActivation activation, @Field(value="hidden") int[] hidden, @Field(value="epochs") double epochs, @Field(value="train_samples_per_iteration") long train_samples_per_iteration, @Field(value="target_ratio_comm_to_comp") double target_ratio_comm_to_comp, @Field(value="seed") long seed, @Field(value="adaptive_rate") boolean adaptive_rate, @Field(value="rho") double rho, @Field(value="epsilon") double epsilon, @Field(value="rate") double rate, @Field(value="rate_annealing") double rate_annealing, @Field(value="rate_decay") double rate_decay, @Field(value="momentum_start") double momentum_start, @Field(value="momentum_ramp") double momentum_ramp, @Field(value="momentum_stable") double momentum_stable, @Field(value="nesterov_accelerated_gradient") boolean nesterov_accelerated_gradient, @Field(value="input_dropout_ratio") double input_dropout_ratio, @Field(value="hidden_dropout_ratios") double[] hidden_dropout_ratios, @Field(value="l1") double l1, @Field(value="l2") double l2, @Field(value="max_w2") float max_w2, @Field(value="initial_weight_distribution") DeepLearningInitialWeightDistribution initial_weight_distribution, @Field(value="initial_weight_scale") double initial_weight_scale, @Field(value="initial_weights") java.lang.String[] initial_weights, @Field(value="initial_biases") java.lang.String[] initial_biases, @Field(value="loss") DeepLearningLoss loss, @Field(value="score_interval") double score_interval, @Field(value="score_training_samples") long score_training_samples, @Field(value="score_validation_samples") long score_validation_samples, @Field(value="score_duty_cycle") double score_duty_cycle, @Field(value="classification_stop") double classification_stop, @Field(value="regression_stop") double regression_stop, @Field(value="quiet_mode") boolean quiet_mode, @Field(value="score_validation_sampling") DeepLearningClassSamplingMethod score_validation_sampling, @Field(value="overwrite_with_best_model") boolean overwrite_with_best_model, @Field(value="autoencoder") boolean autoencoder, @Field(value="use_all_factor_levels") boolean use_all_factor_levels, @Field(value="standardize") boolean standardize, @Field(value="diagnostics") boolean diagnostics, @Field(value="variable_importances") boolean variable_importances, @Field(value="fast_mode") boolean fast_mode, @Field(value="force_load_balance") boolean force_load_balance, @Field(value="replicate_training_data") boolean replicate_training_data, @Field(value="single_node_mode") boolean single_node_mode, @Field(value="shuffle_training_data") boolean shuffle_training_data, @Field(value="missing_values_handling") DeepLearningMissingValuesHandling missing_values_handling, @Field(value="sparse") boolean sparse, @Field(value="col_major") boolean col_major, @Field(value="average_activation") double average_activation, @Field(value="sparsity_beta") double sparsity_beta, @Field(value="max_categorical_features") int max_categorical_features, @Field(value="reproducible") boolean reproducible, @Field(value="export_weights_and_biases") boolean export_weights_and_biases, @Field(value="mini_batch_size") int mini_batch_size, @Field(value="elastic_averaging") boolean elastic_averaging, @Field(value="elastic_averaging_moving_rate") double elastic_averaging_moving_rate, @Field(value="elastic_averaging_regularization") double elastic_averaging_regularization, @Field(value="pretrained_autoencoder") java.lang.String pretrained_autoencoder, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
balance_classes - Balance training data class counts via over/under-sampling (for imbalanced data).class_sampling_factors - Desired over/under-sampling ratios per class (in lexicographic order). If not
specified, sampling factors will be automatically computed to obtain class balance
during training. Requires balance_classes.max_after_balance_size - Maximum relative size of the training data after balancing class counts (can be
less than 1.0). Requires balance_classes.max_confusion_matrix_size - [Deprecated] Maximum size (# classes) for confusion matrices to be printed in
the Logs.max_hit_ratio_k - Max. number (top K) of predictions to use for hit ratio computation (for multi-class only,
0 to disable).activation - Activation function.hidden - Hidden layer sizes (e.g. [100, 100]).epochs - How many times the dataset should be iterated (streamed), can be fractional.train_samples_per_iteration - Number of training samples (globally) per MapReduce iteration. Special values
are 0: one epoch, -1: all available data (e.g., replicated training data), -2:
automatic.target_ratio_comm_to_comp - Target ratio of communication overhead to computation. Only for multi-node
operation and train_samples_per_iteration = -2 (auto-tuning).seed - Seed for random numbers (affects sampling) - Note: only reproducible when running single threaded.adaptive_rate - Adaptive learning rate.rho - Adaptive learning rate time decay factor (similarity to prior updates).epsilon - Adaptive learning rate smoothing factor (to avoid divisions by zero and allow progress).rate - Learning rate (higher => less stable, lower => slower convergence).rate_annealing - Learning rate annealing: rate / (1 + rate_annealing * samples).rate_decay - Learning rate decay factor between layers (N-th layer: rate * rate_decay ^ (n - 1).momentum_start - Initial momentum at the beginning of training (try 0.5).momentum_ramp - Number of training samples for which momentum increases.momentum_stable - Final momentum after the ramp is over (try 0.99).nesterov_accelerated_gradient - Use Nesterov accelerated gradient (recommended).input_dropout_ratio - Input layer dropout ratio (can improve generalization, try 0.1 or 0.2).hidden_dropout_ratios - Hidden layer dropout ratios (can improve generalization), specify one value per
hidden layer, defaults to 0.5.l1 - L1 regularization (can add stability and improve generalization, causes many weights to become 0).l2 - L2 regularization (can add stability and improve generalization, causes many weights to be small.max_w2 - Constraint for squared sum of incoming weights per unit (e.g. for Rectifier).initial_weight_distribution - Initial weight distribution.initial_weight_scale - Uniform: -value...value, Normal: stddev.initial_weights - A list of H2OFrame ids to initialize the weight matrices of this model with.initial_biases - A list of H2OFrame ids to initialize the bias vectors of this model with.loss - Loss function.score_interval - Shortest time interval (in seconds) between model scoring.score_training_samples - Number of training set samples for scoring (0 for all).score_validation_samples - Number of validation set samples for scoring (0 for all).score_duty_cycle - Maximum duty cycle fraction for scoring (lower: more training, higher: more scoring).classification_stop - Stopping criterion for classification error fraction on training data (-1 to disable).regression_stop - Stopping criterion for regression error (MSE) on training data (-1 to disable).quiet_mode - Enable quiet mode for less output to standard output.score_validation_sampling - Method used to sample validation dataset for scoring.overwrite_with_best_model - If enabled, override the final model with the best model found during training.autoencoder - Auto-Encoder.use_all_factor_levels - Use all factor levels of categorical variables. Otherwise, the first factor level is
omitted (without loss of accuracy). Useful for variable importances and auto-enabled
for autoencoder.standardize - If enabled, automatically standardize the data. If disabled, the user must provide properly
scaled input data.diagnostics - Enable diagnostics for hidden layers.variable_importances - Compute variable importances for input features (Gedeon method) - can be slow for
large networks.fast_mode - Enable fast mode (minor approximation in back-propagation).force_load_balance - Force extra load balancing to increase training speed for small datasets (to keep all
cores busy).replicate_training_data - Replicate the entire training dataset onto every node for faster training on small
datasets.single_node_mode - Run on a single node for fine-tuning of model parameters.shuffle_training_data - Enable shuffling of training data (recommended if training data is replicated and
train_samples_per_iteration is close to #nodes x #rows, of if using
balance_classes).missing_values_handling - Handling of missing values. Either MeanImputation or Skip.sparse - Sparse data handling (more efficient for data with lots of 0 values).col_major - #DEPRECATED Use a column major weight matrix for input layer. Can speed up forward propagation,
but might slow down backpropagation.average_activation - Average activation for sparse auto-encoder. #Experimentalsparsity_beta - Sparsity regularization. #Experimentalmax_categorical_features - Max. number of categorical features, enforced via hashing. #Experimentalreproducible - Force reproducibility on small data (will be slow - only uses 1 thread).export_weights_and_biases - Whether to export Neural Network weights and biases to H2O Frames.mini_batch_size - Mini-batch size (smaller leads to better fit, larger can speed up and generalize better).elastic_averaging - Elastic averaging between compute nodes can improve distributed model convergence.
#Experimentalelastic_averaging_moving_rate - Elastic averaging moving rate (only if elastic averaging is enabled).elastic_averaging_regularization - Elastic averaging regularization strength (only if elastic averaging is
enabled).pretrained_autoencoder - Pretrained autoencoder model to initialize this model with.model_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/deeplearning") retrofit2.Call<DeepLearningV3> trainDeeplearning()
@FormUrlEncoded @POST(value="/99/Grid/glm") retrofit2.Call<GLMV3> trainGlm(@Field(value="seed") long seed, @Field(value="family") GLMFamily family, @Field(value="tweedie_variance_power") double tweedie_variance_power, @Field(value="tweedie_link_power") double tweedie_link_power, @Field(value="solver") GLMSolver solver, @Field(value="alpha") double[] alpha, @Field(value="lambda") double[] lambda, @Field(value="lambda_search") boolean lambda_search, @Field(value="early_stopping") boolean early_stopping, @Field(value="nlambdas") int nlambdas, @Field(value="standardize") boolean standardize, @Field(value="missing_values_handling") DeepLearningMissingValuesHandling missing_values_handling, @Field(value="non_negative") boolean non_negative, @Field(value="max_iterations") int max_iterations, @Field(value="beta_epsilon") double beta_epsilon, @Field(value="objective_epsilon") double objective_epsilon, @Field(value="gradient_epsilon") double gradient_epsilon, @Field(value="obj_reg") double obj_reg, @Field(value="link") GLMLink link, @Field(value="intercept") boolean intercept, @Field(value="prior") double prior, @Field(value="lambda_min_ratio") double lambda_min_ratio, @Field(value="beta_constraints") java.lang.String beta_constraints, @Field(value="max_active_predictors") int max_active_predictors, @Field(value="interactions") java.lang.String[] interactions, @Field(value="interaction_pairs") StringPairV3[] interaction_pairs, @Field(value="balance_classes") boolean balance_classes, @Field(value="class_sampling_factors") float[] class_sampling_factors, @Field(value="max_after_balance_size") float max_after_balance_size, @Field(value="max_confusion_matrix_size") int max_confusion_matrix_size, @Field(value="max_hit_ratio_k") int max_hit_ratio_k, @Field(value="compute_p_values") boolean compute_p_values, @Field(value="remove_collinear_columns") boolean remove_collinear_columns, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
seed - Seed for pseudo random number generator (if applicable)family - Family. Use binomial for classification with logistic regression, others are for regression
problems.tweedie_variance_power - Tweedie variance powertweedie_link_power - Tweedie link powersolver - AUTO will set the solver based on given data and the other parameters. IRLSM is fast on on problems
with small number of predictors and for lambda-search with L1 penalty, L_BFGS scales better for
datasets with many columns. Coordinate descent is experimental (beta).alpha - Distribution of regularization between the L1 (Lasso) and L2 (Ridge) penalties. A value of 1 for
alpha represents Lasso regression, a value of 0 produces Ridge regression, and anything in between
specifies the amount of mixing between the two. Default value of alpha is 0 when SOLVER = 'L-BFGS';
0.5 otherwise.lambda - Regularization strengthlambda_search - Use lambda search starting at lambda max, given lambda is then interpreted as lambda minearly_stopping - Stop early when there is no more relative improvement on train or validation (if provided)nlambdas - Number of lambdas to be used in a search. Default indicates: If alpha is zero, with lambda search
set to True, the value of nlamdas is set to 30 (fewer lambdas are needed for ridge regression)
otherwise it is set to 100.standardize - Standardize numeric columns to have zero mean and unit variancemissing_values_handling - Handling of missing values. Either MeanImputation or Skip.non_negative - Restrict coefficients (not intercept) to be non-negativemax_iterations - Maximum number of iterationsbeta_epsilon - Converge if beta changes less (using L-infinity norm) than beta esilon, ONLY applies to
IRLSM solverobjective_epsilon - Converge if objective value changes less than this. Default indicates: If lambda_search
is set to True the value of objective_epsilon is set to .0001. If the lambda_search is
set to False and lambda is equal to zero, the value of objective_epsilon is set to
.000001, for any other value of lambda the default value of objective_epsilon is set to
.0001.gradient_epsilon - Converge if objective changes less (using L-infinity norm) than this, ONLY applies to
L-BFGS solver. Default indicates: If lambda_search is set to False and lambda is equal to
zero, the default value of gradient_epsilon is equal to .000001, otherwise the default
value is .0001. If lambda_search is set to True, the conditional values above are 1E-8
and 1E-6 respectively.obj_reg - Likelihood divider in objective value computation, default is 1/nobslink - intercept - Include constant term in the modelprior - Prior probability for y==1. To be used only for logistic regression iff the data has been sampled
and the mean of response does not reflect reality.lambda_min_ratio - Minimum lambda used in lambda search, specified as a ratio of lambda_max (the smallest
lambda that drives all coefficients to zero). Default indicates: if the number of
observations is greater than the number of variables, then lambda_min_ratio is set to
0.0001; if the number of observations is less than the number of variables, then
lambda_min_ratio is set to 0.01.beta_constraints - Beta constraintsmax_active_predictors - Maximum number of active predictors during computation. Use as a stopping criterion
to prevent expensive model building with many predictors. Default indicates: If the
IRLSM solver is used, the value of max_active_predictors is set to 5000 otherwise it
is set to 100000000.interactions - A list of predictor column indices to interact. All pairwise combinations will be computed
for the list.interaction_pairs - A list of pairwise (first order) column interactions.balance_classes - Balance training data class counts via over/under-sampling (for imbalanced data).class_sampling_factors - Desired over/under-sampling ratios per class (in lexicographic order). If not
specified, sampling factors will be automatically computed to obtain class balance
during training. Requires balance_classes.max_after_balance_size - Maximum relative size of the training data after balancing class counts (can be
less than 1.0). Requires balance_classes.max_confusion_matrix_size - [Deprecated] Maximum size (# classes) for confusion matrices to be printed in
the Logsmax_hit_ratio_k - Maximum number (top K) of predictions to use for hit ratio computation (for multi-class
only, 0 to disable)compute_p_values - Request p-values computation, p-values work only with IRLSM solver and no regularizationremove_collinear_columns - In case of linearly dependent columns, remove some of the dependent columnsmodel_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/glm") retrofit2.Call<GLMV3> trainGlm()
@FormUrlEncoded @POST(value="/99/Grid/glrm") retrofit2.Call<GLRMV3> trainGlrm(@Field(value="transform") DataInfoTransformType transform, @Field(value="k") int k, @Field(value="loss") GenmodelalgosglrmGlrmLoss loss, @Field(value="multi_loss") GenmodelalgosglrmGlrmLoss multi_loss, @Field(value="loss_by_col") GenmodelalgosglrmGlrmLoss[] loss_by_col, @Field(value="loss_by_col_idx") int[] loss_by_col_idx, @Field(value="period") int period, @Field(value="regularization_x") GenmodelalgosglrmGlrmRegularizer regularization_x, @Field(value="regularization_y") GenmodelalgosglrmGlrmRegularizer regularization_y, @Field(value="gamma_x") double gamma_x, @Field(value="gamma_y") double gamma_y, @Field(value="max_iterations") int max_iterations, @Field(value="max_updates") int max_updates, @Field(value="init_step_size") double init_step_size, @Field(value="min_step_size") double min_step_size, @Field(value="seed") long seed, @Field(value="init") GenmodelalgosglrmGlrmInitialization init, @Field(value="svd_method") SVDMethod svd_method, @Field(value="user_y") java.lang.String user_y, @Field(value="user_x") java.lang.String user_x, @Field(value="loading_name") java.lang.String loading_name, @Field(value="expand_user_y") boolean expand_user_y, @Field(value="impute_original") boolean impute_original, @Field(value="recover_svd") boolean recover_svd, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
transform - Transformation of training datak - Rank of matrix approximationloss - Numeric loss functionmulti_loss - Categorical loss functionloss_by_col - Loss function by column (override)loss_by_col_idx - Loss function by column index (override)period - Length of period (only used with periodic loss function)regularization_x - Regularization function for X matrixregularization_y - Regularization function for Y matrixgamma_x - Regularization weight on X matrixgamma_y - Regularization weight on Y matrixmax_iterations - Maximum number of iterationsmax_updates - Maximum number of updates, defaults to 2*max_iterationsinit_step_size - Initial step sizemin_step_size - Minimum step sizeseed - RNG seed for initializationinit - Initialization modesvd_method - Method for computing SVD during initialization (Caution: Randomized is currently experimental
and unstable)user_y - User-specified initial Yuser_x - User-specified initial Xloading_name - Frame key to save resulting Xexpand_user_y - Expand categorical columns in user-specified initial Yimpute_original - Reconstruct original training data by reversing transformrecover_svd - Recover singular values and eigenvectors of XYmodel_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/glrm") retrofit2.Call<GLRMV3> trainGlrm(@Field(value="k") int k)
@FormUrlEncoded @POST(value="/99/Grid/kmeans") retrofit2.Call<KMeansV3> trainKmeans(@Field(value="user_points") java.lang.String user_points, @Field(value="max_iterations") int max_iterations, @Field(value="standardize") boolean standardize, @Field(value="seed") long seed, @Field(value="init") KMeansInitialization init, @Field(value="estimate_k") boolean estimate_k, @Field(value="k") int k, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
user_points - This option allows you to specify a dataframe, where each row represents an initial cluster
center. The user-specified points must have the same number of columns as the training
observations. The number of rows must equal the number of clustersmax_iterations - Maximum training iterations (if estimate_k is enabled, then this is for each inner Lloyds
iteration)standardize - Standardize columns before computing distancesseed - RNG Seedinit - Initialization modeestimate_k - Whether to estimate the number of clusters (<=k) iteratively and deterministically.k - The max. number of clusters. If estimate_k is disabled, the model will find k centroids, otherwise it
will find up to k centroids.model_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/kmeans") retrofit2.Call<KMeansV3> trainKmeans()
@FormUrlEncoded @POST(value="/99/Grid/naivebayes") retrofit2.Call<NaiveBayesV3> trainNaivebayes(@Field(value="balance_classes") boolean balance_classes, @Field(value="class_sampling_factors") float[] class_sampling_factors, @Field(value="max_after_balance_size") float max_after_balance_size, @Field(value="max_confusion_matrix_size") int max_confusion_matrix_size, @Field(value="max_hit_ratio_k") int max_hit_ratio_k, @Field(value="laplace") double laplace, @Field(value="min_sdev") double min_sdev, @Field(value="eps_sdev") double eps_sdev, @Field(value="min_prob") double min_prob, @Field(value="eps_prob") double eps_prob, @Field(value="compute_metrics") boolean compute_metrics, @Field(value="seed") long seed, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
balance_classes - Balance training data class counts via over/under-sampling (for imbalanced data).class_sampling_factors - Desired over/under-sampling ratios per class (in lexicographic order). If not
specified, sampling factors will be automatically computed to obtain class balance
during training. Requires balance_classes.max_after_balance_size - Maximum relative size of the training data after balancing class counts (can be
less than 1.0). Requires balance_classes.max_confusion_matrix_size - [Deprecated] Maximum size (# classes) for confusion matrices to be printed in
the Logsmax_hit_ratio_k - Max. number (top K) of predictions to use for hit ratio computation (for multi-class only,
0 to disable)laplace - Laplace smoothing parametermin_sdev - Min. standard deviation to use for observations with not enough dataeps_sdev - Cutoff below which standard deviation is replaced with min_sdevmin_prob - Min. probability to use for observations with not enough dataeps_prob - Cutoff below which probability is replaced with min_probcompute_metrics - Compute metrics on training dataseed - Seed for pseudo random number generator (only used for cross-validation and fold_assignment="Random"
or "AUTO")model_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/naivebayes") retrofit2.Call<NaiveBayesV3> trainNaivebayes()
@FormUrlEncoded @POST(value="/99/Grid/pca") retrofit2.Call<PCAV3> trainPca(@Field(value="transform") DataInfoTransformType transform, @Field(value="pca_method") PCAMethod pca_method, @Field(value="k") int k, @Field(value="max_iterations") int max_iterations, @Field(value="seed") long seed, @Field(value="use_all_factor_levels") boolean use_all_factor_levels, @Field(value="compute_metrics") boolean compute_metrics, @Field(value="impute_missing") boolean impute_missing, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
transform - Transformation of training datapca_method - Method for computing PCA (Caution: GLRM is currently experimental and unstable)k - Rank of matrix approximationmax_iterations - Maximum training iterationsseed - RNG seed for initializationuse_all_factor_levels - Whether first factor level is included in each categorical expansioncompute_metrics - Whether to compute metrics on the training dataimpute_missing - Whether to impute missing entries with the column meanmodel_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/pca") retrofit2.Call<PCAV3> trainPca(@Field(value="k") int k)
@FormUrlEncoded @POST(value="/99/Grid/svd") retrofit2.Call<SVDV99> trainSvd(@Field(value="transform") DataInfoTransformType transform, @Field(value="svd_method") SVDMethod svd_method, @Field(value="nv") int nv, @Field(value="max_iterations") int max_iterations, @Field(value="seed") long seed, @Field(value="keep_u") boolean keep_u, @Field(value="u_name") java.lang.String u_name, @Field(value="use_all_factor_levels") boolean use_all_factor_levels, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
transform - Transformation of training datasvd_method - Method for computing SVD (Caution: Randomized is currently experimental and unstable)nv - Number of right singular vectorsmax_iterations - Maximum iterationsseed - RNG seed for k-means++ initializationkeep_u - Save left singular vectors?u_name - Frame key to save left singular vectorsuse_all_factor_levels - Whether first factor level is included in each categorical expansionmodel_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/svd") retrofit2.Call<SVDV99> trainSvd()
@FormUrlEncoded @POST(value="/99/Grid/drf") retrofit2.Call<DRFV3> trainDrf(@Field(value="mtries") int mtries, @Field(value="binomial_double_trees") boolean binomial_double_trees, @Field(value="balance_classes") boolean balance_classes, @Field(value="class_sampling_factors") float[] class_sampling_factors, @Field(value="max_after_balance_size") float max_after_balance_size, @Field(value="max_confusion_matrix_size") int max_confusion_matrix_size, @Field(value="max_hit_ratio_k") int max_hit_ratio_k, @Field(value="ntrees") int ntrees, @Field(value="max_depth") int max_depth, @Field(value="min_rows") double min_rows, @Field(value="nbins") int nbins, @Field(value="nbins_top_level") int nbins_top_level, @Field(value="nbins_cats") int nbins_cats, @Field(value="r2_stopping") double r2_stopping, @Field(value="seed") long seed, @Field(value="build_tree_one_node") boolean build_tree_one_node, @Field(value="sample_rate") double sample_rate, @Field(value="sample_rate_per_class") double[] sample_rate_per_class, @Field(value="col_sample_rate_per_tree") double col_sample_rate_per_tree, @Field(value="col_sample_rate_change_per_level") double col_sample_rate_change_per_level, @Field(value="score_tree_interval") int score_tree_interval, @Field(value="min_split_improvement") double min_split_improvement, @Field(value="histogram_type") TreeSharedTreeModelSharedTreeParametersHistogramType histogram_type, @Field(value="calibrate_model") boolean calibrate_model, @Field(value="calibration_frame") java.lang.String calibration_frame, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
mtries - Number of variables randomly sampled as candidates at each split. If set to -1, defaults to sqrt{p}
for classification and p/3 for regression (where p is the # of predictorsbinomial_double_trees - For binary classification: Build 2x as many trees (one per class) - can lead to
higher accuracy.balance_classes - Balance training data class counts via over/under-sampling (for imbalanced data).class_sampling_factors - Desired over/under-sampling ratios per class (in lexicographic order). If not
specified, sampling factors will be automatically computed to obtain class balance
during training. Requires balance_classes.max_after_balance_size - Maximum relative size of the training data after balancing class counts (can be
less than 1.0). Requires balance_classes.max_confusion_matrix_size - [Deprecated] Maximum size (# classes) for confusion matrices to be printed in
the Logsmax_hit_ratio_k - Max. number (top K) of predictions to use for hit ratio computation (for multi-class only,
0 to disable)ntrees - Number of trees.max_depth - Maximum tree depth.min_rows - Fewest allowed (weighted) observations in a leaf.nbins - For numerical columns (real/int), build a histogram of (at least) this many bins, then split at the
best pointnbins_top_level - For numerical columns (real/int), build a histogram of (at most) this many bins at the
root level, then decrease by factor of two per levelnbins_cats - For categorical columns (factors), build a histogram of this many bins, then split at the best
point. Higher values can lead to more overfitting.r2_stopping - r2_stopping is no longer supported and will be ignored if set - please use stopping_rounds,
stopping_metric and stopping_tolerance instead. Previous version of H2O would stop making
trees when the R^2 metric equals or exceeds thisseed - Seed for pseudo random number generator (if applicable)build_tree_one_node - Run on one node only; no network overhead but fewer cpus used. Suitable for small
datasets.sample_rate - Row sample rate per tree (from 0.0 to 1.0)sample_rate_per_class - A list of row sample rates per class (relative fraction for each class, from 0.0 to
1.0), for each treecol_sample_rate_per_tree - Column sample rate per tree (from 0.0 to 1.0)col_sample_rate_change_per_level - Relative change of the column sampling rate for every level (must be >
0.0 and <= 2.0)score_tree_interval - Score the model after every so many trees. Disabled if set to 0.min_split_improvement - Minimum relative improvement in squared error reduction for a split to happenhistogram_type - What type of histogram to use for finding optimal split pointscalibrate_model - Use Platt Scaling to calculate calibrated class probabilities. Calibration can provide
more accurate estimates of class probabilities.calibration_frame - Calibration frame for Platt Scalingmodel_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/drf") retrofit2.Call<DRFV3> trainDrf()
@FormUrlEncoded @POST(value="/99/Grid/gbm") retrofit2.Call<GBMV3> trainGbm(@Field(value="learn_rate") double learn_rate, @Field(value="learn_rate_annealing") double learn_rate_annealing, @Field(value="col_sample_rate") double col_sample_rate, @Field(value="max_abs_leafnode_pred") double max_abs_leafnode_pred, @Field(value="pred_noise_bandwidth") double pred_noise_bandwidth, @Field(value="balance_classes") boolean balance_classes, @Field(value="class_sampling_factors") float[] class_sampling_factors, @Field(value="max_after_balance_size") float max_after_balance_size, @Field(value="max_confusion_matrix_size") int max_confusion_matrix_size, @Field(value="max_hit_ratio_k") int max_hit_ratio_k, @Field(value="ntrees") int ntrees, @Field(value="max_depth") int max_depth, @Field(value="min_rows") double min_rows, @Field(value="nbins") int nbins, @Field(value="nbins_top_level") int nbins_top_level, @Field(value="nbins_cats") int nbins_cats, @Field(value="r2_stopping") double r2_stopping, @Field(value="seed") long seed, @Field(value="build_tree_one_node") boolean build_tree_one_node, @Field(value="sample_rate") double sample_rate, @Field(value="sample_rate_per_class") double[] sample_rate_per_class, @Field(value="col_sample_rate_per_tree") double col_sample_rate_per_tree, @Field(value="col_sample_rate_change_per_level") double col_sample_rate_change_per_level, @Field(value="score_tree_interval") int score_tree_interval, @Field(value="min_split_improvement") double min_split_improvement, @Field(value="histogram_type") TreeSharedTreeModelSharedTreeParametersHistogramType histogram_type, @Field(value="calibrate_model") boolean calibrate_model, @Field(value="calibration_frame") java.lang.String calibration_frame, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
learn_rate - Learning rate (from 0.0 to 1.0)learn_rate_annealing - Scale the learning rate by this factor after each tree (e.g., 0.99 or 0.999)col_sample_rate - Column sample rate (from 0.0 to 1.0)max_abs_leafnode_pred - Maximum absolute value of a leaf node predictionpred_noise_bandwidth - Bandwidth (sigma) of Gaussian multiplicative noise ~N(1,sigma) for tree node
predictionsbalance_classes - Balance training data class counts via over/under-sampling (for imbalanced data).class_sampling_factors - Desired over/under-sampling ratios per class (in lexicographic order). If not
specified, sampling factors will be automatically computed to obtain class balance
during training. Requires balance_classes.max_after_balance_size - Maximum relative size of the training data after balancing class counts (can be
less than 1.0). Requires balance_classes.max_confusion_matrix_size - [Deprecated] Maximum size (# classes) for confusion matrices to be printed in
the Logsmax_hit_ratio_k - Max. number (top K) of predictions to use for hit ratio computation (for multi-class only,
0 to disable)ntrees - Number of trees.max_depth - Maximum tree depth.min_rows - Fewest allowed (weighted) observations in a leaf.nbins - For numerical columns (real/int), build a histogram of (at least) this many bins, then split at the
best pointnbins_top_level - For numerical columns (real/int), build a histogram of (at most) this many bins at the
root level, then decrease by factor of two per levelnbins_cats - For categorical columns (factors), build a histogram of this many bins, then split at the best
point. Higher values can lead to more overfitting.r2_stopping - r2_stopping is no longer supported and will be ignored if set - please use stopping_rounds,
stopping_metric and stopping_tolerance instead. Previous version of H2O would stop making
trees when the R^2 metric equals or exceeds thisseed - Seed for pseudo random number generator (if applicable)build_tree_one_node - Run on one node only; no network overhead but fewer cpus used. Suitable for small
datasets.sample_rate - Row sample rate per tree (from 0.0 to 1.0)sample_rate_per_class - A list of row sample rates per class (relative fraction for each class, from 0.0 to
1.0), for each treecol_sample_rate_per_tree - Column sample rate per tree (from 0.0 to 1.0)col_sample_rate_change_per_level - Relative change of the column sampling rate for every level (must be >
0.0 and <= 2.0)score_tree_interval - Score the model after every so many trees. Disabled if set to 0.min_split_improvement - Minimum relative improvement in squared error reduction for a split to happenhistogram_type - What type of histogram to use for finding optimal split pointscalibrate_model - Use Platt Scaling to calculate calibrated class probabilities. Calibration can provide
more accurate estimates of class probabilities.calibration_frame - Calibration frame for Platt Scalingmodel_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/gbm") retrofit2.Call<GBMV3> trainGbm()
@FormUrlEncoded @POST(value="/99/Grid/aggregator") retrofit2.Call<AggregatorV99> trainAggregator(@Field(value="transform") DataInfoTransformType transform, @Field(value="pca_method") PCAMethod pca_method, @Field(value="k") int k, @Field(value="max_iterations") int max_iterations, @Field(value="target_num_exemplars") int target_num_exemplars, @Field(value="rel_tol_num_exemplars") double rel_tol_num_exemplars, @Field(value="seed") long seed, @Field(value="use_all_factor_levels") boolean use_all_factor_levels, @Field(value="save_mapping_frame") boolean save_mapping_frame, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
transform - Transformation of training datapca_method - Method for computing PCA (Caution: GLRM is currently experimental and unstable)k - Rank of matrix approximationmax_iterations - Maximum number of iterations for PCAtarget_num_exemplars - Targeted number of exemplarsrel_tol_num_exemplars - Relative tolerance for number of exemplars (e.g, 0.5 is +/- 50 percents)seed - RNG seed for initializationuse_all_factor_levels - Whether first factor level is included in each categorical expansionsave_mapping_frame - Whether to export the mapping of the aggregated framemodel_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/aggregator") retrofit2.Call<AggregatorV99> trainAggregator()
@FormUrlEncoded @POST(value="/99/Grid/deepwater") retrofit2.Call<DeepWaterV3> trainDeepwater(@Field(value="problem_type") DeepWaterParametersProblemType problem_type, @Field(value="activation") DeepWaterParametersActivation activation, @Field(value="hidden") int[] hidden, @Field(value="input_dropout_ratio") double input_dropout_ratio, @Field(value="hidden_dropout_ratios") double[] hidden_dropout_ratios, @Field(value="max_confusion_matrix_size") int max_confusion_matrix_size, @Field(value="sparse") boolean sparse, @Field(value="max_hit_ratio_k") int max_hit_ratio_k, @Field(value="epochs") double epochs, @Field(value="train_samples_per_iteration") long train_samples_per_iteration, @Field(value="target_ratio_comm_to_comp") double target_ratio_comm_to_comp, @Field(value="seed") long seed, @Field(value="learning_rate") double learning_rate, @Field(value="learning_rate_annealing") double learning_rate_annealing, @Field(value="momentum_start") double momentum_start, @Field(value="momentum_ramp") double momentum_ramp, @Field(value="momentum_stable") double momentum_stable, @Field(value="score_interval") double score_interval, @Field(value="score_training_samples") long score_training_samples, @Field(value="score_validation_samples") long score_validation_samples, @Field(value="score_duty_cycle") double score_duty_cycle, @Field(value="classification_stop") double classification_stop, @Field(value="regression_stop") double regression_stop, @Field(value="quiet_mode") boolean quiet_mode, @Field(value="overwrite_with_best_model") boolean overwrite_with_best_model, @Field(value="autoencoder") boolean autoencoder, @Field(value="diagnostics") boolean diagnostics, @Field(value="variable_importances") boolean variable_importances, @Field(value="replicate_training_data") boolean replicate_training_data, @Field(value="single_node_mode") boolean single_node_mode, @Field(value="shuffle_training_data") boolean shuffle_training_data, @Field(value="mini_batch_size") int mini_batch_size, @Field(value="clip_gradient") double clip_gradient, @Field(value="network") DeepWaterParametersNetwork network, @Field(value="backend") DeepWaterParametersBackend backend, @Field(value="image_shape") int[] image_shape, @Field(value="channels") int channels, @Field(value="gpu") boolean gpu, @Field(value="device_id") int[] device_id, @Field(value="cache_data") boolean cache_data, @Field(value="network_definition_file") java.lang.String network_definition_file, @Field(value="network_parameters_file") java.lang.String network_parameters_file, @Field(value="mean_image_file") java.lang.String mean_image_file, @Field(value="export_native_parameters_prefix") java.lang.String export_native_parameters_prefix, @Field(value="standardize") boolean standardize, @Field(value="balance_classes") boolean balance_classes, @Field(value="class_sampling_factors") float[] class_sampling_factors, @Field(value="max_after_balance_size") float max_after_balance_size, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
problem_type - Problem type, auto-detected by default. If set to image, the H2OFrame must contain a string
column containing the path (URI or URL) to the images in the first column. If set to text,
the H2OFrame must contain a string column containing the text in the first column. If set to
dataset, Deep Water behaves just like any other H2O Model and builds a model on the provided
H2OFrame (non-String columns).activation - Activation function. Only used if no user-defined network architecture file is provided, and
only for problem_type=dataset.hidden - Hidden layer sizes (e.g. [200, 200]). Only used if no user-defined network architecture file is
provided, and only for problem_type=dataset.input_dropout_ratio - Input layer dropout ratio (can improve generalization, try 0.1 or 0.2).hidden_dropout_ratios - Hidden layer dropout ratios (can improve generalization), specify one value per
hidden layer, defaults to 0.5.max_confusion_matrix_size - [Deprecated] Maximum size (# classes) for confusion matrices to be printed in
the Logs.sparse - Sparse data handling (more efficient for data with lots of 0 values).max_hit_ratio_k - Max. number (top K) of predictions to use for hit ratio computation (for multi-class only,
0 to disable).epochs - How many times the dataset should be iterated (streamed), can be fractional.train_samples_per_iteration - Number of training samples (globally) per MapReduce iteration. Special values
are 0: one epoch, -1: all available data (e.g., replicated training data), -2:
automatic.target_ratio_comm_to_comp - Target ratio of communication overhead to computation. Only for multi-node
operation and train_samples_per_iteration = -2 (auto-tuning).seed - Seed for random numbers (affects sampling) - Note: only reproducible when running single threaded.learning_rate - Learning rate (higher => less stable, lower => slower convergence).learning_rate_annealing - Learning rate annealing: rate / (1 + rate_annealing * samples).momentum_start - Initial momentum at the beginning of training (try 0.5).momentum_ramp - Number of training samples for which momentum increases.momentum_stable - Final momentum after the ramp is over (try 0.99).score_interval - Shortest time interval (in seconds) between model scoring.score_training_samples - Number of training set samples for scoring (0 for all).score_validation_samples - Number of validation set samples for scoring (0 for all).score_duty_cycle - Maximum duty cycle fraction for scoring (lower: more training, higher: more scoring).classification_stop - Stopping criterion for classification error fraction on training data (-1 to disable).regression_stop - Stopping criterion for regression error (MSE) on training data (-1 to disable).quiet_mode - Enable quiet mode for less output to standard output.overwrite_with_best_model - If enabled, override the final model with the best model found during training.autoencoder - Auto-Encoder.diagnostics - Enable diagnostics for hidden layers.variable_importances - Compute variable importances for input features (Gedeon method) - can be slow for
large networks.replicate_training_data - Replicate the entire training dataset onto every node for faster training on small
datasets.single_node_mode - Run on a single node for fine-tuning of model parameters.shuffle_training_data - Enable global shuffling of training data.mini_batch_size - Mini-batch size (smaller leads to better fit, larger can speed up and generalize better).clip_gradient - Clip gradients once their absolute value is larger than this value.network - Network architecture.backend - Deep Learning Backend.image_shape - Width and height of image.channels - Number of (color) channels.gpu - Whether to use a GPU (if available).device_id - Device IDs (which GPUs to use).cache_data - Whether to cache the data in memory (automatically disabled if data size is too large).network_definition_file - Path of file containing network definition (graph, architecture).network_parameters_file - Path of file containing network (initial) parameters (weights, biases).mean_image_file - Path of file containing the mean image data for data normalization.export_native_parameters_prefix - Path (prefix) where to export the native model parameters after every
iteration.standardize - If enabled, automatically standardize the data. If disabled, the user must provide properly
scaled input data.balance_classes - Balance training data class counts via over/under-sampling (for imbalanced data).class_sampling_factors - Desired over/under-sampling ratios per class (in lexicographic order). If not
specified, sampling factors will be automatically computed to obtain class balance
during training. Requires balance_classes.max_after_balance_size - Maximum relative size of the training data after balancing class counts (can be
less than 1.0). Requires balance_classes.model_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/deepwater") retrofit2.Call<DeepWaterV3> trainDeepwater()
@FormUrlEncoded @POST(value="/99/Grid/word2vec") retrofit2.Call<Word2VecV3> trainWord2vec(@Field(value="vec_size") int vec_size, @Field(value="window_size") int window_size, @Field(value="sent_sample_rate") float sent_sample_rate, @Field(value="norm_model") Word2VecNormModel norm_model, @Field(value="epochs") int epochs, @Field(value="min_word_freq") int min_word_freq, @Field(value="init_learning_rate") float init_learning_rate, @Field(value="word_model") Word2VecWordModel word_model, @Field(value="pre_trained") java.lang.String pre_trained, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
vec_size - Set size of word vectorswindow_size - Set max skip length between wordssent_sample_rate - Set threshold for occurrence of words. Those that appear with higher frequency in the
training data will be randomly down-sampled; useful range is (0, 1e-5)norm_model - Use Hierarchical Softmaxepochs - Number of training iterations to runmin_word_freq - This will discard words that appear less than init_learning_rate - Set the starting learning rateword_model - Use the Skip-Gram modelpre_trained - Id of a data frame that contains a pre-trained (external) word2vec modelmodel_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/word2vec") retrofit2.Call<Word2VecV3> trainWord2vec()
@FormUrlEncoded @POST(value="/99/Grid/stackedensemble") retrofit2.Call<StackedEnsembleV99> trainStackedensemble(@Field(value="base_models") java.lang.String[] base_models, @Field(value="metalearner_algorithm") StackedEnsembleModelStackedEnsembleParametersMetalearnerAlgorithm metalearner_algorithm, @Field(value="metalearner_nfolds") int metalearner_nfolds, @Field(value="metalearner_fold_assignment") ModelParametersFoldAssignmentScheme metalearner_fold_assignment, @Field(value="metalearner_fold_column") java.lang.String metalearner_fold_column, @Field(value="keep_levelone_frame") boolean keep_levelone_frame, @Field(value="metalearner_params") java.lang.String metalearner_params, @Field(value="seed") long seed, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
base_models - List of models (or model ids) to ensemble/stack together. Models must have been cross-
validated using nfolds > 1, and folds must be identical across models.metalearner_algorithm - Type of algorithm to use as the metalearner. Options include 'AUTO' (GLM with non
negative weights; if validation_frame is present, a lambda search is performed),
'glm' (GLM with default parameters), 'gbm' (GBM with default parameters), 'drf'
(Random Forest with default parameters), or 'deeplearning' (Deep Learning with
default parameters).metalearner_nfolds - Number of folds for K-fold cross-validation of the metalearner algorithm (0 to disable
or >= 2).metalearner_fold_assignment - Cross-validation fold assignment scheme for metalearner cross-validation.
Defaults to AUTO (which is currently set to Random). The 'Stratified' option
will stratify the folds based on the response variable, for classification
problems.metalearner_fold_column - Column with cross-validation fold index assignment per observation for cross-
validation of the metalearner.keep_levelone_frame - Keep level one frame used for metalearner training.metalearner_params - Parameters for metalearner algorithmseed - Seed for random numbers; passed through to the metalearner algorithm. Defaults to -1 (time-based
random number)model_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/stackedensemble") retrofit2.Call<StackedEnsembleV99> trainStackedensemble(@Field(value="base_models") java.lang.String[] base_models)
@FormUrlEncoded @POST(value="/99/Grid/coxph") retrofit2.Call<CoxPHV3> trainCoxph(@Field(value="rcall") java.lang.String rcall, @Field(value="stop_column") java.lang.String stop_column, @Field(value="start_column") java.lang.String start_column, @Field(value="ties") CoxPHTies ties, @Field(value="init") double init, @Field(value="lre_min") double lre_min, @Field(value="iter_max") int iter_max, @Field(value="model_id") java.lang.String model_id, @Field(value="training_frame") java.lang.String training_frame, @Field(value="validation_frame") java.lang.String validation_frame, @Field(value="nfolds") int nfolds, @Field(value="keep_cross_validation_predictions") boolean keep_cross_validation_predictions, @Field(value="keep_cross_validation_fold_assignment") boolean keep_cross_validation_fold_assignment, @Field(value="parallelize_cross_validation") boolean parallelize_cross_validation, @Field(value="distribution") GenmodelutilsDistributionFamily distribution, @Field(value="tweedie_power") double tweedie_power, @Field(value="quantile_alpha") double quantile_alpha, @Field(value="huber_alpha") double huber_alpha, @Field(value="response_column") java.lang.String response_column, @Field(value="weights_column") java.lang.String weights_column, @Field(value="offset_column") java.lang.String offset_column, @Field(value="fold_column") java.lang.String fold_column, @Field(value="fold_assignment") ModelParametersFoldAssignmentScheme fold_assignment, @Field(value="categorical_encoding") ModelParametersCategoricalEncodingScheme categorical_encoding, @Field(value="max_categorical_levels") int max_categorical_levels, @Field(value="ignored_columns") java.lang.String[] ignored_columns, @Field(value="ignore_const_cols") boolean ignore_const_cols, @Field(value="score_each_iteration") boolean score_each_iteration, @Field(value="checkpoint") java.lang.String checkpoint, @Field(value="stopping_rounds") int stopping_rounds, @Field(value="max_runtime_secs") double max_runtime_secs, @Field(value="stopping_metric") ScoreKeeperStoppingMetric stopping_metric, @Field(value="stopping_tolerance") double stopping_tolerance, @Field(value="custom_metric_func") java.lang.String custom_metric_func)
rcall - rcallstop_column - stop_columnstart_column - start_columnties - tiesinit - initlre_min - lre_miniter_max - iter_maxmodel_id - Destination id for this model; auto-generated if not specified.training_frame - Id of the training data frame.validation_frame - Id of the validation data frame.nfolds - Number of folds for K-fold cross-validation (0 to disable or >= 2).keep_cross_validation_predictions - Whether to keep the predictions of the cross-validation models.keep_cross_validation_fold_assignment - Whether to keep the cross-validation fold assignment.parallelize_cross_validation - Allow parallel training of cross-validation modelsdistribution - Distribution functiontweedie_power - Tweedie power for Tweedie regression, must be between 1 and 2.quantile_alpha - Desired quantile for Quantile regression, must be between 0 and 1.huber_alpha - Desired quantile for Huber/M-regression (threshold between quadratic and linear loss, must be
between 0 and 1).response_column - Response variable column.weights_column - Column with observation weights. Giving some observation a weight of zero is equivalent to
excluding it from the dataset; giving an observation a relative weight of 2 is equivalent
to repeating that row twice. Negative weights are not allowed. Note: Weights are per-row
observation weights and do not increase the size of the data frame. This is typically the
number of times a row is repeated, but non-integer values are supported as well. During
training, rows with higher weights matter more, due to the larger loss function pre-factor.offset_column - Offset column. This will be added to the combination of columns before applying the link
function.fold_column - Column with cross-validation fold index assignment per observation.fold_assignment - Cross-validation fold assignment scheme, if fold_column is not specified. The 'Stratified'
option will stratify the folds based on the response variable, for classification
problems.categorical_encoding - Encoding scheme for categorical featuresmax_categorical_levels - For every categorical feature, only use this many most frequent categorical levels
for model training. Only used for categorical_encoding == EnumLimited.ignored_columns - Names of columns to ignore for training.ignore_const_cols - Ignore constant columns.score_each_iteration - Whether to score during each iteration of model training.checkpoint - Model checkpoint to resume training with.stopping_rounds - Early stopping based on convergence of stopping_metric. Stop if simple moving average of
length k of the stopping_metric does not improve for k:=stopping_rounds scoring events (0
to disable)max_runtime_secs - Maximum allowed runtime in seconds for model training. Use 0 to disable.stopping_metric - Metric to use for early stopping (AUTO: logloss for classification, deviance for
regression)stopping_tolerance - Relative tolerance for metric-based stopping criterion (stop if relative improvement is
not at least this much)custom_metric_func - Reference to custom evaluation function, format: `language:keyName=funcName`@FormUrlEncoded @POST(value="/99/Grid/coxph") retrofit2.Call<CoxPHV3> trainCoxph()