GLMParametersV3 (h2o-java-rest-bindings version 3.8.0.5 API)

java.lang.Object
- water.bindings.pojos.Schema
- - water.bindings.pojos.ModelParametersSchema
  - - water.bindings.pojos.GLMParametersV3

public class GLMParametersV3
extends ModelParametersSchema

Field Summary

Fields
Modifier and Type	Field and Description
`double[]`	`alpha` distribution of regularization between L1 and L2.
`boolean`	`balance_classes` Balance training data class counts via over/under-sampling (for imbalanced data).
`java.lang.String`	`beta_constraints` beta constraints
`double`	`beta_epsilon` converge if beta changes less (using L-infinity norm) than beta esilon, ONLY applies to IRLSM solver
`java.lang.String`	`checkpoint` Model checkpoint to resume training with
`float[]`	`class_sampling_factors` Desired over/under-sampling ratios per class (in lexicographic order).
`boolean`	`compute_p_values` request p-values computation, p-values work only with IRLSM solver and no regularization
`Family`	`family` Family.
`FoldAssignmentScheme`	`fold_assignment` Cross-validation fold assignment scheme, if fold_column is not specified
`ColSpecifierV3`	`fold_column` Column with cross-validation fold index assignment per observation
`double`	`gradient_epsilon` converge if objective changes less (using L-infinity norm) than this, ONLY applies to L-BFGS solver
`boolean`	`ignore_const_cols` Ignore constant columns
`java.lang.String[]`	`ignored_columns` Ignored columns
`boolean`	`intercept` include constant term in the model
`boolean`	`keep_cross_validation_predictions` Keep cross-validation model predictions
`double[]`	`lambda` regularization strength
`double`	`lambda_min_ratio` min lambda used in lambda search, specified as a ratio of lambda_max
`boolean`	`lambda_search` use lambda search starting at lambda max, given lambda is then interpreted as lambda min
`Link`	`link`
`int`	`max_active_predictors` Maximum number of active predictors during computation.
`float`	`max_after_balance_size` Maximum relative size of the training data after balancing class counts (can be less than 1.0).
`int`	`max_confusion_matrix_size` Maximum size (# classes) for confusion matrices to be printed in the Logs
`int`	`max_hit_ratio_k` Max.
`int`	`max_iterations` Maximum number of iterations
`double`	`max_runtime_secs` Maximum allowed runtime in seconds for model training.
`MissingValuesHandling`	`missing_values_handling` Handling of missing values.
`java.lang.String`	`model_id` Destination id for this model; auto-generated if not specified
`int`	`nfolds` Number of folds for N-fold cross-validation
`int`	`nlambdas` number of lambdas to be used in a search
`boolean`	`non_negative` Restrict coefficients (not intercept) to be non-negative
`double`	`obj_reg` likelihood divider in objective value computation, default is 1/nobs
`double`	`objective_epsilon` converge if objective value changes less than this
`ColSpecifierV3`	`offset_column` Offset column
`boolean`	`parallelize_cross_validation` Allow parallel training of cross-validation models
`double`	`prior` prior probability for y==1.
`boolean`	`remove_collinear_columns` in case of linearly dependent columns remove some of the dependent columns
`ColSpecifierV3`	`response_column` Response column
`boolean`	`score_each_iteration` Whether to score during each iteration of model training
`Solver`	`solver` AUTO will set the solver based on given data and the other parameters.
`boolean`	`standardize` Standardize numeric columns to have zero mean and unit variance
`StoppingMetric`	`stopping_metric` Metric to use for early stopping (AUTO: logloss for classification, deviance for regression)
`int`	`stopping_rounds` Early stopping based on convergence of stopping_metric.
`double`	`stopping_tolerance` Relative tolerance for metric-based stopping criterion Relative tolerance for metric-based stopping criterion (stop if relative improvement is not at least this much)
`java.lang.String`	`training_frame` Training frame
`double`	`tweedie_link_power` Tweedie link power
`double`	`tweedie_variance_power` Tweedie variance power
`java.lang.String`	`validation_frame` Validation frame
`ColSpecifierV3`	`weights_column` Column with observation weights

Constructor Summary

Constructors
Constructor and Description

GLMParametersV3()

Constructors
Constructor and Description
`GLMParametersV3()`

Method Summary
- Methods inherited from class java.lang.Object
  clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - family
```
public Family family
```
    Family. Use binomial for classification with logistic regression, others are for regression problems.
  - tweedie_variance_power
```
public double tweedie_variance_power
```
    Tweedie variance power
  - tweedie_link_power
```
public double tweedie_link_power
```
    Tweedie link power
  - solver
```
public Solver solver
```
    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
```
public double[] alpha
```
    distribution of regularization between L1 and L2.
  - lambda
```
public double[] lambda
```
    regularization strength
  - lambda_search
```
public boolean lambda_search
```
    use lambda search starting at lambda max, given lambda is then interpreted as lambda min
  - nlambdas
```
public int nlambdas
```
    number of lambdas to be used in a search
  - standardize
```
public boolean standardize
```
    Standardize numeric columns to have zero mean and unit variance
  - missing_values_handling
```
public MissingValuesHandling missing_values_handling
```
    Handling of missing values. Either Skip or MeanImputation.
  - non_negative
```
public boolean non_negative
```
    Restrict coefficients (not intercept) to be non-negative
  - max_iterations
```
public int max_iterations
```
    Maximum number of iterations
  - beta_epsilon
```
public double beta_epsilon
```
    converge if beta changes less (using L-infinity norm) than beta esilon, ONLY applies to IRLSM solver
  - objective_epsilon
```
public double objective_epsilon
```
    converge if objective value changes less than this
  - gradient_epsilon
```
public double gradient_epsilon
```
    converge if objective changes less (using L-infinity norm) than this, ONLY applies to L-BFGS solver
  - obj_reg
```
public double obj_reg
```
    likelihood divider in objective value computation, default is 1/nobs
  - link
```
public Link link
```
  - intercept
```
public boolean intercept
```
    include constant term in the model
  - prior
```
public double prior
```
    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
```
public double lambda_min_ratio
```
    min lambda used in lambda search, specified as a ratio of lambda_max
  - beta_constraints
```
public java.lang.String beta_constraints
```
    beta constraints
  - max_active_predictors
```
public int max_active_predictors
```
    Maximum number of active predictors during computation. Use as a stopping criterium to prevent expensive model building with many predictors.
  - balance_classes
```
public boolean balance_classes
```
    Balance training data class counts via over/under-sampling (for imbalanced data).
  - class_sampling_factors
```
public float[] 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
```
public float 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
```
public int max_confusion_matrix_size
```
    Maximum size (# classes) for confusion matrices to be printed in the Logs
  - max_hit_ratio_k
```
public int max_hit_ratio_k
```
    Max. number (top K) of predictions to use for hit ratio computation (for multi-class only, 0 to disable)
  - compute_p_values
```
public boolean compute_p_values
```
    request p-values computation, p-values work only with IRLSM solver and no regularization
  - remove_collinear_columns
```
public boolean remove_collinear_columns
```
    in case of linearly dependent columns remove some of the dependent columns
  - model_id
```
public java.lang.String model_id
```
    Destination id for this model; auto-generated if not specified
  - training_frame
```
public java.lang.String training_frame
```
    Training frame
  - validation_frame
```
public java.lang.String validation_frame
```
    Validation frame
  - nfolds
```
public int nfolds
```
    Number of folds for N-fold cross-validation
  - keep_cross_validation_predictions
```
public boolean keep_cross_validation_predictions
```
    Keep cross-validation model predictions
  - parallelize_cross_validation
```
public boolean parallelize_cross_validation
```
    Allow parallel training of cross-validation models
  - response_column
```
public ColSpecifierV3 response_column
```
    Response column
  - weights_column
```
public ColSpecifierV3 weights_column
```
    Column with observation weights
  - offset_column
```
public ColSpecifierV3 offset_column
```
    Offset column
  - fold_column
```
public ColSpecifierV3 fold_column
```
    Column with cross-validation fold index assignment per observation
  - fold_assignment
```
public FoldAssignmentScheme fold_assignment
```
    Cross-validation fold assignment scheme, if fold_column is not specified
  - ignored_columns
```
public java.lang.String[] ignored_columns
```
    Ignored columns
  - ignore_const_cols
```
public boolean ignore_const_cols
```
    Ignore constant columns
  - score_each_iteration
```
public boolean score_each_iteration
```
    Whether to score during each iteration of model training
  - checkpoint
```
public java.lang.String checkpoint
```
    Model checkpoint to resume training with
  - stopping_rounds
```
public int 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
```
public double max_runtime_secs
```
    Maximum allowed runtime in seconds for model training. Use 0 to disable.
  - stopping_metric
```
public StoppingMetric stopping_metric
```
    Metric to use for early stopping (AUTO: logloss for classification, deviance for regression)
  - stopping_tolerance
```
public double stopping_tolerance
```
    Relative tolerance for metric-based stopping criterion Relative tolerance for metric-based stopping criterion (stop if relative improvement is not at least this much)
- Constructor Detail
  - GLMParametersV3
```
public GLMParametersV3()
```

Class GLMParametersV3

Field Summary

Constructor Summary

Method Summary

Methods inherited from class java.lang.Object

Field Detail

family

tweedie_variance_power

tweedie_link_power

solver

alpha

lambda

lambda_search

nlambdas

standardize

missing_values_handling

non_negative

max_iterations

beta_epsilon

objective_epsilon

gradient_epsilon

obj_reg

link

intercept

prior

lambda_min_ratio

beta_constraints

max_active_predictors

balance_classes

class_sampling_factors

max_after_balance_size

max_confusion_matrix_size

max_hit_ratio_k

compute_p_values

remove_collinear_columns

model_id

training_frame

validation_frame

nfolds

keep_cross_validation_predictions

parallelize_cross_validation

response_column

weights_column

offset_column

fold_column

fold_assignment

ignored_columns

ignore_const_cols

score_each_iteration

checkpoint

stopping_rounds

max_runtime_secs

stopping_metric

stopping_tolerance

Constructor Detail

GLMParametersV3