Package hex.deeplearning

Interface Summary

Interface	Description
Storage.Matrix	Abstract matrix interface
Storage.Tensor	Abstract tensor interface
Storage.Vector	Abstract vector interface

Class Summary

Class	Description
DeepLearning	Deep Learning Neural Net implementation based on MRTask
DeepLearningModel	The Deep Learning model It contains a DeepLearningModelInfo with the most up-to-date model, a scoring history, as well as some helpers to indicate the progress
DeepLearningModel.DeepLearningModelOutput	The Deep Learning model output contains a few extra fields in addition to the metrics in Model.Output 1) Scoring history (raw data) 2) weights/biases (raw data) 3) variable importances (TwoDimTable)
DeepLearningModel.DeepLearningParameters	Deep Learning Parameters
DeepLearningModelInfo	This class contains the state of the Deep Learning model This will be shared: one per node
DeepLearningModelInfo.GradientCheck
DeepLearningMojoWriter
DeepLearningScoringInfo	Lightweight DeepLearning scoring history.
DeepLearningTask
DeepLearningTask2	DRemoteTask-based Deep Learning.
Dropout	Helper class for dropout training of Neural Nets
MurmurHash	This is a very fast, non-cryptographic hash suitable for general hash-based lookup.
Neurons	This class implements the concept of a Neuron layer in a Neural Network During training, every MRTask F/J thread is expected to create these neurons for every map call (Cheap to make).
Neurons.ExpRectifier
Neurons.ExpRectifierDropout	Exponential Rectifier with dropout
Neurons.Input	Input layer of the Neural Network This layer is different from other layers as it has no incoming weights, but instead gets its activation values from the training points.
Neurons.Linear	Output neurons for regression - Linear units
Neurons.Maxout	Maxout neurons (picks the max out of the k activation_j = sum(A_ij*x_i) + b_j) Requires k times the model parameters (weights/biases) as a "normal" neuron
Neurons.MaxoutDropout	Maxout neurons with dropout
Neurons.Output	Abstract class for Output neurons
Neurons.Rectifier	Rectifier linear unit (ReLU) neurons
Neurons.RectifierDropout	Rectifier linear unit (ReLU) neurons with dropout
Neurons.Softmax	Output neurons for classification - Softmax
Neurons.Tanh	Tanh neurons - most common, most stable
Neurons.TanhDropout	Tanh neurons with dropout
Storage
Storage.DenseColMatrix	Dense column matrix implementation
Storage.DenseRowMatrix	Dense row matrix implementation
Storage.DenseVector	Dense vector implementation
Storage.SparseRowMatrix	Sparse row matrix implementation

Enum Summary

Enum	Description
DeepLearningModel.DeepLearningParameters.Activation	Activation functions
DeepLearningModel.DeepLearningParameters.ClassSamplingMethod
DeepLearningModel.DeepLearningParameters.InitialWeightDistribution
DeepLearningModel.DeepLearningParameters.Loss	Loss functions Absolute, Quadratic, Huber, Quantile for regression Quadratic, ModifiedHuber or CrossEntropy for classification
DeepLearningModel.DeepLearningParameters.MissingValuesHandling