| Class | Description |
|---|---|
| BaseStatsTask |
Class for storing and updating basic column stats (max, min, mean, sigma).
|
| ColSummaryTask | |
| ConfusionMatrix | |
| Covariance |
Calculate the covariance and correlation of two variables
|
| Covariance.COV_Task | |
| CreateFrame |
Create a Frame from scratch
If randomize = true, then the frame is filled with Random values.
|
| DGLM | |
| DGLM.FamilyIced |
passthrough class around family that properly supports icing
|
| DGLM.GLMJob | |
| DGLM.GLMModel | |
| DGLM.GLMModel.GLMValidationTask | |
| DGLM.GLMParams | |
| DGLM.GLMValidation | |
| DGLM.GLMValidationFunc | |
| DGLM.GramMatrixFunc | |
| DGLM.LambdaMax | |
| DGLM.LambdaMaxFunc | |
| DGLM.LinkIced |
passthrough class around Link that supports Icing
|
| DLSM |
Distributed least squares solvers
|
| DLSM.ADMMSolver | |
| DLSM.GeneralizedGradientSolver |
Generalized gradient solver for solving LSM problem with combination of L1 and L2 penalty.
|
| DLSM.LSMSolver | |
| FrameSplitter |
Frame splitter function to divide given frame into
multiple partitions based on given ratios.
|
| FrameTask<T extends FrameTask<T>> | |
| FrameTask.DataInfo | |
| GLMGrid | |
| GLMGrid.GLMModels | |
| GridSearch | |
| GridSearch.GridSearchProgress | |
| Histogram | |
| Histogram.BinningTask | |
| Histogram.Bins | |
| Histogram.OutlineTask | |
| KMeans |
Scalable K-Means++ (KMeans||)
http://theory.stanford.edu/~sergei/papers/vldb12-kmpar.pdf http://www.youtube.com/watch?v=cigXAxV3XcY |
| KMeans.Lloyds | |
| KMeans.Sampler | |
| KMeans.Sqr | |
| KMeans2 |
Scalable K-Means++ (KMeans||)
http://theory.stanford.edu/~sergei/papers/vldb12-kmpar.pdf http://www.youtube.com/watch?v=cigXAxV3XcY |
| KMeans2.KMeans2Model | |
| KMeans2.KMeans2ModelView | |
| KMeans2.KMeans2Progress | |
| KMeans2.Lloyds | |
| KMeans2.Sampler | |
| KMeans2.SumSqr | |
| KMeansModel | |
| KMeansModel.KMeansApply | |
| KMeansModel.KMeansScore | |
| KMeansShared | |
| Layer |
Neural network layer.
|
| Layer.Input | |
| Layer.Linear |
Linear output layer is used for regression
Rows with missing values in the response column will be ignored
|
| Layer.Maxout | |
| Layer.MaxoutDropout | |
| Layer.Output | |
| Layer.Rectifier | |
| Layer.RectifierDropout | |
| Layer.RectifierPrime | |
| Layer.Softmax |
Softmax output layer is used for classification
Rows with missing values in the response column will be ignored
|
| Layer.Tanh | |
| Layer.TanhDropout | |
| Layer.TanhPrime |
Apply tanh to the weights' transpose.
|
| Layer.VecLinear | |
| Layer.VecsInput | |
| Layer.VecSoftmax | |
| LinearRegression | |
| LinearRegression.CalcRegressionTask | |
| LinearRegression.CalcSquareErrorsTasks | |
| LinearRegression.CalcSumsTask | |
| LinearRegression.LRResult | |
| LR2 | |
| LR2.CalcRegressionTask | |
| LR2.CalcSquareErrorsTasks | |
| LR2.CalcSumsTask | |
| NeuralNet |
Neural network.
|
| NeuralNet.Errors | |
| NeuralNet.NeuralNetModel | |
| NeuralNet.NeuralNetScore | |
| NewRowVecTask<T extends Iced> | |
| NewRowVecTask.DataFrame |
Struct to keep info about our data.
|
| NewRowVecTask.RowFunc<T extends Iced> | |
| NodeShuffle |
Shuffle the rows of some dataset, such that the natural placement of the
resulting ValueArray onto Nodes results in some good property.
|
| NOPTask | |
| OneHot | |
| ParamsSearch |
Looks for parameters on a set of objects and perform random search.
|
| Quantiles |
Quantile of a column.
|
| Quantiles.BinTask2 | |
| ReBalance |
Rebalance a Frame
|
| RowTask<T extends Freezable> | |
| RowTask.Row<T extends Freezable> | |
| RowTask.RowFunction<T extends Iced> | |
| RowVecTask | |
| RowVecTask.Sampling | |
| ScoreTask | |
| ShuffleTask |
Simple shuffle task based on Fisher/Yates algo.
|
| Summary | |
| Summary.ColSummary | |
| Summary2 |
Summary of a column.
|
| Summary2.BasicStat | |
| Summary2.PrePass | |
| Summary2.SummaryPerRow | |
| Summary2.SummaryTask2 | |
| Trainer |
Trains a neural network.
|
| Trainer.Base | |
| Trainer.Direct |
Trains NN on current thread.
|
| Trainer.MapReduce |
Distributed trainer.
|
| Trainer.OpenCL |
GPU based trainer.
|
| Trainer.Threaded |
Runs several trainers in parallel on the same weights, using threads.
|
| VarImp | |
| VarImp.VarImpMDA |
Variable importance measured as mean decrease in accuracy.
|
| VarImp.VarImpRI |
Variable importance measured as relative influence.
|
| Enum | Description |
|---|---|
| ConfusionMatrix.ErrMetric | |
| DGLM.CaseMode | |
| DGLM.Family | |
| DGLM.GLMModel.Status | |
| DGLM.Link | |
| DLSM.LSMSolverType | |
| KMeans.Initialization | |
| NeuralNet.Activation |
Activation functions
|
| NeuralNet.ExecutionMode | |
| NeuralNet.InitialWeightDistribution | |
| NeuralNet.Loss |
Loss functions
CrossEntropy is recommended
|
| RowVecTask.DataPreprocessing |
| Exception | Description |
|---|---|
| DGLM.GLMException | |
| DLSM.ADMMSolver.NonSPDMatrixException | |
| DLSM.LSMSolver.LSMSolverException |
| Annotation Type | Description |
|---|---|
| ParamsSearch.Ignore | |
| ParamsSearch.Info |