public static final class DeepLearningV3.DeepLearningParametersV3 extends water.api.schemas3.ModelParametersSchemaV3<DeepLearningModel.DeepLearningParameters,DeepLearningV3.DeepLearningParametersV3>
Modifier and Type  Field and Description 

DeepLearningModel.DeepLearningParameters.Activation 
activation
The activation function (nonlinearity) to be used by the neurons in the hidden layers.

boolean 
adaptive_rate
The implemented adaptive learning rate algorithm (ADADELTA) automatically
combines the benefits of learning rate annealing and momentum
training to avoid slow convergence.

boolean 
autoencoder 
double 
average_activation 
boolean 
balance_classes
For imbalanced data, balance training data class counts via
over/undersampling.

float[] 
class_sampling_factors
Desired over/undersampling ratios per class (lexicographic order).

double 
classification_stop
The stopping criteria in terms of classification error (1accuracy) on the
training data scoring dataset.

boolean 
col_major 
boolean 
diagnostics
Gather diagnostics for hidden layers, such as mean and RMS values of learning
rate, momentum, weights and biases.

boolean 
elastic_averaging 
double 
elastic_averaging_moving_rate 
double 
elastic_averaging_regularization 
double 
epochs
The number of passes over the training dataset to be carried out.

double 
epsilon
The second of two hyper parameters for adaptive learning rate (ADADELTA).

boolean 
export_weights_and_biases 
boolean 
fast_mode
Enable fast mode (minor approximation in backpropagation), should not affect results significantly.

static java.lang.String[] 
fields 
boolean 
force_load_balance
Increase training speed on small datasets by splitting it into many chunks
to allow utilization of all cores.

int[] 
hidden
The number and size of each hidden layer in the model.

double[] 
hidden_dropout_ratios
A fraction of the inputs for each hidden layer to be omitted from training in order
to improve generalization.

water.api.schemas3.KeyV3.FrameKeyV3[] 
initial_biases 
DeepLearningModel.DeepLearningParameters.InitialWeightDistribution 
initial_weight_distribution
The distribution from which initial weights are to be drawn.

double 
initial_weight_scale
The scale of the distribution function for Uniform or Normal distributions.

water.api.schemas3.KeyV3.FrameKeyV3[] 
initial_weights 
double 
input_dropout_ratio
A fraction of the features for each training row to be omitted from training in order
to improve generalization (dimension sampling).

double 
l1
A regularization method that constrains the absolute value of the weights and
has the net effect of dropping some weights (setting them to zero) from a model
to reduce complexity and avoid overfitting.

double 
l2
A regularization method that constrains the sum of the squared
weights.

DeepLearningModel.DeepLearningParameters.Loss 
loss
The loss (error) function to be minimized by the model.

float 
max_after_balance_size
When classes are balanced, limit the resulting dataset size to the
specified multiple of the original dataset size.

int 
max_categorical_features 
int 
max_confusion_matrix_size
For classification models, the maximum size (in terms of classes) of
the confusion matrix for it to be printed.

int 
max_hit_ratio_k
The maximum number (top K) of predictions to use for hit ratio computation (for multiclass only, 0 to disable)

float 
max_w2
A maximum on the sum of the squared incoming weights into
any one neuron.

int 
mini_batch_size 
DeepLearningModel.DeepLearningParameters.MissingValuesHandling 
missing_values_handling 
double 
momentum_ramp
The momentum_ramp parameter controls the amount of learning for which momentum increases
(assuming momentum_stable is larger than momentum_start).

double 
momentum_stable
The momentum_stable parameter controls the final momentum value reached after momentum_ramp training samples.

double 
momentum_start
The momentum_start parameter controls the amount of momentum at the beginning of training.

boolean 
nesterov_accelerated_gradient
The Nesterov accelerated gradient descent method is a modification to
traditional gradient descent for convex functions.

boolean 
overwrite_with_best_model
If enabled, store the best model under the destination key of this model at the end of training.

water.api.schemas3.KeyV3.ModelKeyV3 
pretrained_autoencoder 
boolean 
quiet_mode
Enable quiet mode for less output to standard output.

double 
rate
When adaptive learning rate is disabled, the magnitude of the weight
updates are determined by the user specified learning rate
(potentially annealed), and are a function of the difference
between the predicted value and the target value.

double 
rate_annealing
Learning rate annealing reduces the learning rate to "freeze" into
local minima in the optimization landscape.

double 
rate_decay
The learning rate decay parameter controls the change of learning rate across layers.

double 
regression_stop
The stopping criteria in terms of regression error (MSE) on the training
data scoring dataset.

boolean 
replicate_training_data
Replicate the entire training dataset onto every node for faster training on small datasets.

boolean 
reproducible 
double 
rho
The first of two hyper parameters for adaptive learning rate (ADADELTA).

double 
score_duty_cycle
Maximum fraction of wall clock time spent on model scoring on training and validation samples,
and on diagnostics such as computation of feature importances (i.e., not on training).

double 
score_interval
The minimum time (in seconds) to elapse between model scoring.

long 
score_training_samples
The number of training dataset points to be used for scoring.

long 
score_validation_samples
The number of validation dataset points to be used for scoring.

DeepLearningModel.DeepLearningParameters.ClassSamplingMethod 
score_validation_sampling
Method used to sample the validation dataset for scoring, see Score Validation Samples above.

long 
seed
The random seed controls sampling and initialization.

boolean 
shuffle_training_data
Enable shuffling of training data (on each node).

boolean 
single_node_mode
Run on a single node for finetuning of model parameters.

boolean 
sparse 
double 
sparsity_beta 
boolean 
standardize 
double 
target_ratio_comm_to_comp 
long 
train_samples_per_iteration
The number of training data rows to be processed per iteration.

boolean 
use_all_factor_levels 
boolean 
variable_importances
Whether to compute variable importances for input features.

categorical_encoding, checkpoint, custom_distribution_func, custom_metric_func, distribution, export_checkpoints_dir, fold_assignment, fold_column, huber_alpha, ignore_const_cols, ignored_columns, keep_cross_validation_fold_assignment, keep_cross_validation_models, keep_cross_validation_predictions, max_categorical_levels, max_runtime_secs, model_id, nfolds, offset_column, parallelize_cross_validation, quantile_alpha, response_column, score_each_iteration, stopping_metric, stopping_rounds, stopping_tolerance, training_frame, tweedie_power, validation_frame, weights_column
Constructor and Description 

DeepLearningParametersV3() 
append_field_arrays, fields, fillFromImpl, fillImpl, writeParametersJSON
createAndFillImpl, createImpl, extractVersionFromSchemaName, fillFromBody, fillFromImpl, fillFromImpl, fillFromParms, fillFromParms, fillImpl, getImplClass, getImplClass, getSchemaName, getSchemaType, getSchemaVersion, init_meta, markdown, markdown, newInstance, newInstance, setField, setSchemaType_doNotCall
public static java.lang.String[] fields
@API(level=secondary, direction=INOUT, gridable=true, help="Balance training data class counts via over/undersampling (for imbalanced data).") public boolean balance_classes
@API(level=expert, direction=INOUT, gridable=true, help="Desired over/undersampling ratios per class (in lexicographic order). If not specified, sampling factors will be automatically computed to obtain class balance during training. Requires balance_classes.") public float[] class_sampling_factors
@API(level=expert, direction=INOUT, gridable=false, help="Maximum relative size of the training data after balancing class counts (can be less than 1.0). Requires balance_classes.") public float max_after_balance_size
@API(level=secondary, direction=INOUT, gridable=false, help="[Deprecated] Maximum size (# classes) for confusion matrices to be printed in the Logs.") public int max_confusion_matrix_size
@API(level=secondary, direction=INOUT, gridable=true, help="Max. number (top K) of predictions to use for hit ratio computation (for multiclass only, 0 to disable).") public int max_hit_ratio_k
@API(level=critical, direction=INOUT, gridable=true, values={"Tanh","TanhWithDropout","Rectifier","RectifierWithDropout","Maxout","MaxoutWithDropout"}, help="Activation function.") public DeepLearningModel.DeepLearningParameters.Activation activation
@API(level=critical, direction=INOUT, gridable=true, help="Hidden layer sizes (e.g. [100, 100]).") public int[] hidden
@API(level=critical, direction=INOUT, gridable=true, help="How many times the dataset should be iterated (streamed), can be fractional.") public double epochs
@API(level=secondary, direction=INOUT, gridable=true, help="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.") public long train_samples_per_iteration
@API(level=expert, direction=INOUT, gridable=true, help="Target ratio of communication overhead to computation. Only for multinode operation and train_samples_per_iteration = 2 (autotuning).") public double target_ratio_comm_to_comp
@API(level=expert, direction=INOUT, gridable=true, help="Seed for random numbers (affects sampling)  Note: only reproducible when running single threaded.") public long seed
@API(level=secondary, direction=INOUT, gridable=true, help="Adaptive learning rate.") public boolean adaptive_rate
@API(level=expert, direction=INOUT, gridable=true, help="Adaptive learning rate time decay factor (similarity to prior updates).") public double rho
@API(level=expert, direction=INOUT, gridable=true, help="Adaptive learning rate smoothing factor (to avoid divisions by zero and allow progress).") public double epsilon
@API(level=expert, direction=INOUT, gridable=true, help="Learning rate (higher => less stable, lower => slower convergence).") public double rate
@API(level=expert, direction=INOUT, gridable=true, help="Learning rate annealing: rate / (1 + rate_annealing * samples).") public double rate_annealing
@API(level=expert, direction=INOUT, gridable=true, help="Learning rate decay factor between layers (Nth layer: rate * rate_decay ^ (n  1).") public double rate_decay
@API(level=expert, direction=INOUT, gridable=true, help="Initial momentum at the beginning of training (try 0.5).") public double momentum_start
@API(level=expert, direction=INOUT, gridable=true, help="Number of training samples for which momentum increases.") public double momentum_ramp
@API(level=expert, direction=INOUT, gridable=true, help="Final momentum after the ramp is over (try 0.99).") public double momentum_stable
@API(level=expert, direction=INOUT, gridable=true, help="Use Nesterov accelerated gradient (recommended).") public boolean nesterov_accelerated_gradient
@API(level=secondary, direction=INOUT, gridable=true, help="Input layer dropout ratio (can improve generalization, try 0.1 or 0.2).") public double input_dropout_ratio
@API(level=secondary, direction=INOUT, gridable=true, help="Hidden layer dropout ratios (can improve generalization), specify one value per hidden layer, defaults to 0.5.") public double[] hidden_dropout_ratios
@API(level=secondary, direction=INOUT, gridable=true, help="L1 regularization (can add stability and improve generalization, causes many weights to become 0).") public double l1
@API(level=secondary, direction=INOUT, gridable=true, help="L2 regularization (can add stability and improve generalization, causes many weights to be small.") public double l2
@API(level=expert, direction=INOUT, gridable=true, help="Constraint for squared sum of incoming weights per unit (e.g. for Rectifier).") public float max_w2
@API(level=expert, direction=INOUT, gridable=true, values={"UniformAdaptive","Uniform","Normal"}, help="Initial weight distribution.") public DeepLearningModel.DeepLearningParameters.InitialWeightDistribution initial_weight_distribution
@API(level=expert, direction=INOUT, gridable=true, help="Uniform: value...value, Normal: stddev.") public double initial_weight_scale
@API(level=expert, direction=INOUT, gridable=true, help="A list of H2OFrame ids to initialize the weight matrices of this model with.") public water.api.schemas3.KeyV3.FrameKeyV3[] initial_weights
@API(level=expert, direction=INOUT, gridable=true, help="A list of H2OFrame ids to initialize the bias vectors of this model with.") public water.api.schemas3.KeyV3.FrameKeyV3[] initial_biases
@API(level=secondary, direction=INOUT, gridable=true, required=false, values={"Automatic","CrossEntropy","Quadratic","Huber","Absolute","Quantile"}, help="Loss function.") public DeepLearningModel.DeepLearningParameters.Loss loss
@API(level=secondary, direction=INOUT, gridable=true, help="Shortest time interval (in seconds) between model scoring.") public double score_interval
@API(level=secondary, direction=INOUT, gridable=true, help="Number of training set samples for scoring (0 for all).") public long score_training_samples
@API(level=secondary, direction=INOUT, gridable=true, help="Number of validation set samples for scoring (0 for all).") public long score_validation_samples
@API(level=secondary, direction=INOUT, gridable=true, help="Maximum duty cycle fraction for scoring (lower: more training, higher: more scoring).") public double score_duty_cycle
@API(level=expert, direction=INOUT, gridable=true, help="Stopping criterion for classification error fraction on training data (1 to disable).") public double classification_stop
@API(level=expert, direction=INOUT, gridable=true, help="Stopping criterion for regression error (MSE) on training data (1 to disable).") public double regression_stop
@API(level=expert, direction=INOUT, gridable=true, help="Enable quiet mode for less output to standard output.") public boolean quiet_mode
@API(level=expert, direction=INOUT, gridable=true, values={"Uniform","Stratified"}, help="Method used to sample validation dataset for scoring.") public DeepLearningModel.DeepLearningParameters.ClassSamplingMethod score_validation_sampling
@API(level=expert, direction=INOUT, gridable=true, help="If enabled, override the final model with the best model found during training.") public boolean overwrite_with_best_model
@API(level=secondary, direction=INOUT, help="AutoEncoder.") public boolean autoencoder
@API(level=secondary, direction=INOUT, gridable=true, help="Use all factor levels of categorical variables. Otherwise, the first factor level is omitted (without loss of accuracy). Useful for variable importances and autoenabled for autoencoder.") public boolean use_all_factor_levels
@API(level=secondary, direction=INOUT, gridable=true, help="If enabled, automatically standardize the data. If disabled, the user must provide properly scaled input data.") public boolean standardize
@API(level=expert, direction=INOUT, help="Enable diagnostics for hidden layers.") public boolean diagnostics
@API(level=critical, direction=INOUT, gridable=true, help="Compute variable importances for input features (Gedeon method)  can be slow for large networks.") public boolean variable_importances
@API(level=expert, direction=INOUT, gridable=true, help="Enable fast mode (minor approximation in backpropagation).") public boolean fast_mode
@API(level=expert, direction=INOUT, gridable=true, help="Force extra load balancing to increase training speed for small datasets (to keep all cores busy).") public boolean force_load_balance
@API(level=secondary, direction=INOUT, gridable=true, help="Replicate the entire training dataset onto every node for faster training on small datasets.") public boolean replicate_training_data
@API(level=expert, direction=INOUT, gridable=true, help="Run on a single node for finetuning of model parameters.") public boolean single_node_mode
@API(level=expert, direction=INOUT, gridable=true, help="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).") public boolean shuffle_training_data
@API(level=expert, direction=INOUT, gridable=true, values={"MeanImputation","Skip"}, help="Handling of missing values. Either MeanImputation or Skip.") public DeepLearningModel.DeepLearningParameters.MissingValuesHandling missing_values_handling
@API(level=expert, direction=INOUT, gridable=true, help="Sparse data handling (more efficient for data with lots of 0 values).") public boolean sparse
@API(level=expert, direction=INOUT, gridable=true, help="#DEPRECATED Use a column major weight matrix for input layer. Can speed up forward propagation, but might slow down backpropagation.") public boolean col_major
@API(level=expert, direction=INOUT, gridable=true, help="Average activation for sparse autoencoder. #Experimental") public double average_activation
@API(level=expert, direction=INOUT, gridable=true, help="Sparsity regularization. #Experimental") public double sparsity_beta
@API(level=expert, direction=INOUT, gridable=true, help="Max. number of categorical features, enforced via hashing. #Experimental") public int max_categorical_features
@API(level=expert, direction=INOUT, gridable=true, help="Force reproducibility on small data (will be slow  only uses 1 thread).") public boolean reproducible
@API(level=expert, direction=INOUT, help="Whether to export Neural Network weights and biases to H2O Frames.") public boolean export_weights_and_biases
@API(level=expert, direction=INOUT, help="Minibatch size (smaller leads to better fit, larger can speed up and generalize better).") public int mini_batch_size
@API(level=expert, direction=INOUT, gridable=true, help="Elastic averaging between compute nodes can improve distributed model convergence. #Experimental") public boolean elastic_averaging
@API(level=expert, direction=INOUT, gridable=true, help="Elastic averaging moving rate (only if elastic averaging is enabled).") public double elastic_averaging_moving_rate
@API(level=expert, direction=INOUT, gridable=true, help="Elastic averaging regularization strength (only if elastic averaging is enabled).") public double elastic_averaging_regularization
@API(level=expert, direction=INOUT, help="Pretrained autoencoder model to initialize this model with.") public water.api.schemas3.KeyV3.ModelKeyV3 pretrained_autoencoder