Model.Parameters (h2o-core version 3.12.0.1 API)

java.lang.Object
- water.Iced<Model.Parameters>
- - hex.Model.Parameters

All Implemented Interfaces:

java.io.Externalizable, java.io.Serializable, java.lang.Cloneable, Freezable<Model.Parameters>

Direct Known Subclasses:

ClusteringModel.ClusteringParameters, QuantileModel.QuantileParameters

Enclosing class:

Model<M extends Model<M,P,O>,P extends Model.Parameters,O extends Model.Output>
```
public abstract static class Model.Parameters
extends Iced<Model.Parameters>
```
Model-specific parameter class. Each model sub-class contains instance of one of these containing its builder parameters, with model-specific parameters. E.g. KMeansModel extends Model and has a KMeansParameters extending Model.Parameters; sample parameters include K, whether or not to normalize, max iterations and the initial random seed.
The non-transient fields are input parameters to the model-building process, and are considered "first class citizens" by the front-end - the front-end will cache Parameters (in the browser, in JavaScript, on disk) and rebuild Parameter instances from those caches. WARNING: Model Parameters is not immutable object and ModelBuilder can modify them!

See Also:
Serialized Form

Nested Class Summary

Nested Classes
Modifier and Type Class and Description

static class Model.Parameters.CategoricalEncodingScheme

static class Model.Parameters.FoldAssignmentScheme

Nested Classes
Modifier and Type	Class and Description
`static class`	`Model.Parameters.CategoricalEncodingScheme`
`static class`	`Model.Parameters.FoldAssignmentScheme`

Field Summary

Fields
Modifier and Type	Field and Description
`boolean`	`_auto_rebalance`
`boolean`	`_balance_classes` Should all classes be over/under-sampled to balance the class distribution?
`Model.Parameters.CategoricalEncodingScheme`	`_categorical_encoding`
`Key<? extends Model>`	`_checkpoint` A model key associated with a previously trained Deep Learning model.
`float[]`	`_class_sampling_factors` Desired over/under-sampling ratios per class (lexicographic order).
`hex.genmodel.utils.DistributionFamily`	`_distribution`
`Model.Parameters.FoldAssignmentScheme`	`_fold_assignment`
`java.lang.String`	`_fold_column`
`double`	`_huber_alpha`
`boolean`	`_ignore_const_cols`
`java.lang.String[]`	`_ignored_columns`
`boolean`	`_is_cv_model`
`boolean`	`_keep_cross_validation_fold_assignment`
`boolean`	`_keep_cross_validation_predictions`
`float`	`_max_after_balance_size` When classes are being balanced, limit the resulting dataset size to the specified multiple of the original dataset size.
`int`	`_max_categorical_levels`
`int`	`_max_confusion_matrix_size` For classification models, the maximum size (in terms of classes) of the confusion matrix for it to be printed.
`double`	`_max_runtime_secs` Maximum allowed runtime in seconds for model training.
`int`	`_nfolds`
`java.lang.String`	`_offset_column`
`boolean`	`_parallelize_cross_validation`
`Key<? extends Model>`	`_pretrained_autoencoder` A pretrained Autoencoder DL model with matching inputs and hidden layers can be used to initialize the weights and biases (excluding the output layer).
`double`	`_quantile_alpha`
`java.lang.String`	`_response_column` Supervised models have an expected response they get to train with!
`boolean`	`_score_each_iteration`
`long`	`_seed`
`ScoreKeeper.StoppingMetric`	`_stopping_metric` Metric to use for convergence checking, only for _stopping_rounds > 0.
`int`	`_stopping_rounds` Early stopping based on convergence of stopping_metric.
`double`	`_stopping_tolerance` Relative tolerance for metric-based stopping criterion: stop if relative improvement is not at least this much.
`Key<Frame>`	`_train`
`double`	`_tweedie_power`
`Key<Frame>`	`_valid`
`java.lang.String`	`_weights_column`
`static int`	`MAX_SUPPORTED_LEVELS` Maximal number of supported levels in response.

Constructor Summary

Constructors
Constructor and Description

Model.Parameters()

Constructors
Constructor and Description
`Model.Parameters()`

Method Summary

Methods
Modifier and Type	Method and Description
`abstract java.lang.String`	`algoName()` The short name, used in making Keys.
`protected long`	`checksum_impl()` Compute a checksum based on all non-transient non-static ice-able assignable fields (incl.
`long`	`checksum()`
`protected boolean`	`defaultDropConsCols()`
`protected double`	`defaultStoppingTolerance()` Default relative tolerance for convergence-based early stopping
`abstract java.lang.String`	`fullName()` The pretty algo name for this Model (e.g., Gradient Boosting Machine, rather than GBM).
`long`	`getOrMakeRealSeed()`
`boolean`	`hasCheckpoint()`
`abstract java.lang.String`	`javaName()` The Java class name for this Model (e.g., hex.tree.gbm.GBM, rather than GBM).
`double`	`missingColumnsType()` Type of missing columns during adaptation between train/test datasets Overload this method for models that have sparse data handling - a zero will preserve the sparseness.
`abstract long`	`progressUnits()` How much work will be done for this model?
`void`	`read_lock_frames(Job job)` Read-Lock both training and validation User frames.
`void`	`read_unlock_frames(Job job)` Read-UnLock both training and validation User frames.
`void`	`setTrain(Key<Frame> train)`
`Frame`	`train()`
`Frame`	`valid()`

Methods inherited from class water.Iced
asBytes, clone, copyOver, frozenType, read, readExternal, readJSON, reloadFromBytes, toJsonString, write, writeExternal, writeJSON

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

- Field Detail
  - MAX_SUPPORTED_LEVELS
```
public static final int MAX_SUPPORTED_LEVELS
```
    Maximal number of supported levels in response.
    
    See Also:
    Constant Field Values
  - _train
```
public Key<Frame> _train
```
  - _valid
```
public Key<Frame> _valid
```
  - _nfolds
```
public int _nfolds
```
  - _keep_cross_validation_predictions
```
public boolean _keep_cross_validation_predictions
```
  - _keep_cross_validation_fold_assignment
```
public boolean _keep_cross_validation_fold_assignment
```
  - _parallelize_cross_validation
```
public boolean _parallelize_cross_validation
```
  - _auto_rebalance
```
public boolean _auto_rebalance
```
  - _seed
```
public long _seed
```
  - _fold_assignment
```
public Model.Parameters.FoldAssignmentScheme _fold_assignment
```
  - _categorical_encoding
```
public Model.Parameters.CategoricalEncodingScheme _categorical_encoding
```
  - _max_categorical_levels
```
public int _max_categorical_levels
```
  - _distribution
```
public hex.genmodel.utils.DistributionFamily _distribution
```
  - _tweedie_power
```
public double _tweedie_power
```
  - _quantile_alpha
```
public double _quantile_alpha
```
  - _huber_alpha
```
public double _huber_alpha
```
  - _ignored_columns
```
public java.lang.String[] _ignored_columns
```
  - _ignore_const_cols
```
public boolean _ignore_const_cols
```
  - _weights_column
```
public java.lang.String _weights_column
```
  - _offset_column
```
public java.lang.String _offset_column
```
  - _fold_column
```
public java.lang.String _fold_column
```
  - _is_cv_model
```
public boolean _is_cv_model
```
  - _score_each_iteration
```
public boolean _score_each_iteration
```
  - _max_runtime_secs
```
public double _max_runtime_secs
```
    Maximum allowed runtime in seconds for model training. Use 0 to disable.
  - _stopping_rounds
```
public int _stopping_rounds
```
    Early stopping based on convergence of stopping_metric. Stop if simple moving average of the stopping_metric does not improve by stopping_tolerance for k scoring events. Can only trigger after at least 2k scoring events. Use 0 to disable.
  - _stopping_metric
```
public ScoreKeeper.StoppingMetric _stopping_metric
```
    Metric to use for convergence checking, only for _stopping_rounds > 0.
  - _stopping_tolerance
```
public double _stopping_tolerance
```
    Relative tolerance for metric-based stopping criterion: stop if relative improvement is not at least this much.
  - _response_column
```
public java.lang.String _response_column
```
    Supervised models have an expected response they get to train with!
  - _balance_classes
```
public boolean _balance_classes
```
    Should all classes be over/under-sampled to balance the class distribution?
  - _max_after_balance_size
```
public float _max_after_balance_size
```
    When classes are being balanced, limit the resulting dataset size to the specified multiple of the original dataset size. Maximum relative size of the training data after balancing class counts (can be less than 1.0)
  - _class_sampling_factors
```
public float[] _class_sampling_factors
```
    Desired over/under-sampling ratios per class (lexicographic order). Only when balance_classes is enabled. If not specified, they will be automatically computed to obtain class balance during training.
  - _max_confusion_matrix_size
```
public int _max_confusion_matrix_size
```
    For classification models, the maximum size (in terms of classes) of the confusion matrix for it to be printed. This option is meant to avoid printing extremely large confusion matrices.
  - _checkpoint
```
public Key<? extends Model> _checkpoint
```
    A model key associated with a previously trained Deep Learning model. This option allows users to build a new model as a continuation of a previously generated model.
  - _pretrained_autoencoder
```
public Key<? extends Model> _pretrained_autoencoder
```
    A pretrained Autoencoder DL model with matching inputs and hidden layers can be used to initialize the weights and biases (excluding the output layer).
- Constructor Detail
  - Model.Parameters
```
public Model.Parameters()
```
- Method Detail
  - algoName
```
public abstract java.lang.String algoName()
```
    The short name, used in making Keys. e.g. "GBM"
  - fullName
```
public abstract java.lang.String fullName()
```
    The pretty algo name for this Model (e.g., Gradient Boosting Machine, rather than GBM).
  - javaName
```
public abstract java.lang.String javaName()
```
    The Java class name for this Model (e.g., hex.tree.gbm.GBM, rather than GBM).
  - defaultStoppingTolerance
```
protected double defaultStoppingTolerance()
```
    Default relative tolerance for convergence-based early stopping
  - progressUnits
```
public abstract long progressUnits()
```
    How much work will be done for this model?
  - setTrain
```
public void setTrain(Key<Frame> train)
```
  - getOrMakeRealSeed
```
public long getOrMakeRealSeed()
```
  - train
```
public final Frame train()
```
    Returns:
    the training frame instance
  - valid
```
public final Frame valid()
```
    Returns:
    the validation frame instance, or null if a validation frame was not specified
  - read_lock_frames
```
public void read_lock_frames(Job job)
```
    Read-Lock both training and validation User frames.
  - read_unlock_frames
```
public void read_unlock_frames(Job job)
```
    Read-UnLock both training and validation User frames. This method is called on crashing cleanup pathes, so handles the case where the frames are not actually locked.
  - defaultDropConsCols
```
protected boolean defaultDropConsCols()
```
  - missingColumnsType
```
public double missingColumnsType()
```
    Type of missing columns during adaptation between train/test datasets Overload this method for models that have sparse data handling - a zero will preserve the sparseness. Otherwise, NaN is used.
    
    Returns:
    real-valued number (can be NaN)
  - hasCheckpoint
```
public boolean hasCheckpoint()
```
  - checksum
```
public long checksum()
```
  - checksum_impl
```
protected long checksum_impl()
```
    Compute a checksum based on all non-transient non-static ice-able assignable fields (incl. inherited ones) which have @API annotations. Sort the fields first, since reflection gives us the fields in random order and we don't want the checksum to be affected by the field order. NOTE: if a field is added to a Parameters class the checksum will differ even when all the previous parameters have the same value. If a client wants backward compatibility they will need to compare parameter values explicitly. The method is motivated by standard hash implementation `hash = hash * P + value` but we use high prime numbers in random order.
    
    Returns:
    checksum

Class Model.Parameters

Nested Class Summary

Field Summary

Constructor Summary

Method Summary

Methods inherited from class water.Iced

Methods inherited from class java.lang.Object

Field Detail

MAX_SUPPORTED_LEVELS

_train

_valid

_nfolds

_keep_cross_validation_predictions

_keep_cross_validation_fold_assignment

_parallelize_cross_validation

_auto_rebalance

_seed

_fold_assignment

_categorical_encoding

_max_categorical_levels

_distribution

_tweedie_power

_quantile_alpha

_huber_alpha

_ignored_columns

_ignore_const_cols

_weights_column

_offset_column

_fold_column

_is_cv_model

_score_each_iteration

_max_runtime_secs

_stopping_rounds

_stopping_metric

_stopping_tolerance

_response_column

_balance_classes

_max_after_balance_size

_class_sampling_factors

_max_confusion_matrix_size

_checkpoint

_pretrained_autoencoder

Constructor Detail

Model.Parameters

Method Detail

algoName

fullName

javaName

defaultStoppingTolerance

progressUnits

setTrain

getOrMakeRealSeed

train

valid

read_lock_frames

read_unlock_frames

defaultDropConsCols

missingColumnsType

hasCheckpoint

checksum

checksum_impl