| Package | Description |
|---|---|
| hex |
The hex package contains math and machine learning algorithms for H2O.
|
| hex.deeplearning | |
| hex.gbm |
Gradient Boosting Machine.
|
| hex.glm | |
| hex.singlenoderf | |
| water |
The water package contains the implementation of H2O's core.
|
| water.api |
API classes that help implement the embedded web browser interface.
|
| water.exec |
Implementation of the interpreter available in the H2O console.
|
| water.fvec |
Fluid Vector API implementation.
|
| water.util |
Miscellaneous utility classes (for example, for logging).
|
| Modifier and Type | Field and Description |
|---|---|
Vec |
CoxPH.event_column |
Vec |
CoxPH.start_column |
Vec |
CoxPH.stop_column |
Vec |
Layer.VecSoftmax.vec |
Vec[] |
Layer.VecsInput.vecs |
Vec |
CoxPH.weights_column |
| Modifier and Type | Method and Description |
|---|---|
protected Vec[][] |
FrameExtractor.makeTemplates()
Create a templates for vector composing output frame
|
static Vec |
ShuffleTask.shuffle(Vec ivec) |
| Modifier and Type | Method and Description |
|---|---|
protected MRTask2 |
NFoldFrameExtractor.createNewWorker(H2O.H2OCountedCompleter completer,
Vec[] inputVecs,
int split) |
protected abstract MRTask2 |
FrameExtractor.createNewWorker(H2O.H2OCountedCompleter completer,
Vec[] inputVecs,
int split)
Create a new worker which has to setup given completer.
|
static NeuralNet.Errors |
NeuralNet.eval(Layer[] ls,
Vec[] vecs,
Vec resp,
long n,
long[][] cm) |
static NeuralNet.Errors |
NeuralNet.eval(Layer[] ls,
Vec[] vecs,
Vec resp,
long n,
long[][] cm) |
void |
Summary2.finishUp(Vec vec) |
void |
Quantiles.finishUp(Vec vec,
double[] quantiles_to_do,
int interpolation_type,
boolean multiPass) |
void |
Layer.VecsInput.Init(Vec[] vecs,
Layer.VecsInput train) |
static Frame |
FrameTask.DataInfo.prepareFrame(Frame source,
Vec[] response,
int[] ignored_cols,
boolean toEnum,
boolean dropConstantCols,
boolean dropNACols) |
static Frame |
FrameTask.DataInfo.prepareFrame(Frame source,
Vec response,
int[] ignored_cols,
boolean toEnum,
boolean dropConstantCols) |
static Frame |
FrameTask.DataInfo.prepareFrame(Frame source,
Vec response,
int[] ignored_cols,
boolean toEnum,
boolean dropConstantCols,
boolean dropNACols)
Prepare a Frame (with a single response) to be processed by the FrameTask
1) Place response at the end
2) (Optionally) Remove columns with constant values or with greater than 20% NaNs
3) Possibly turn integer categoricals into enums
|
static void |
NeuralNet.reChunk(Vec[] vecs)
Makes sure small datasets are spread over enough chunks to parallelize training.
|
static Vec |
ShuffleTask.shuffle(Vec ivec) |
void |
Summary2.toHTML(Vec vec,
String cname,
StringBuilder sb) |
| Constructor and Description |
|---|
Layer.VecLinear(Vec vec,
Layer.VecLinear stats) |
Layer.VecsInput(Vec[] vecs,
Layer.VecsInput train) |
Layer.VecSoftmax(Vec vec,
Layer.VecSoftmax stats) |
Quantiles(Vec vec) |
Quantiles(Vec vec,
int max_qbins,
double valStart,
double valEnd) |
Summary2(Vec vec,
String name,
Summary2.BasicStat stat0) |
Summary2(Vec vec,
String name,
Summary2.BasicStat stat0,
int max_qbins) |
| Modifier and Type | Method and Description |
|---|---|
double |
DeepLearningModel.calcOutlierThreshold(Vec mse,
double quantile)
Compute quantile-based threshold (in reconstruction error) to find outliers
|
| Modifier and Type | Method and Description |
|---|---|
protected Vec |
SharedTreeModelBuilder.vec_nids(Frame fr,
int t) |
protected Vec |
SharedTreeModelBuilder.vec_resp(Frame fr,
int t) |
protected Vec |
SharedTreeModelBuilder.vec_tree(Frame fr,
int c) |
| Modifier and Type | Field and Description |
|---|---|
Vec |
GLM2.offset |
Vec |
GLM2.Source.offset |
Vec |
GLM2.Source.response |
| Constructor and Description |
|---|
GLM2.Source(Frame fr,
Vec response,
boolean standardize) |
GLM2.Source(Frame fr,
Vec response,
boolean standardize,
boolean intercept) |
GLM2.Source(Frame fr,
Vec response,
boolean standardize,
boolean intercept,
Vec offset) |
| Modifier and Type | Method and Description |
|---|---|
static VariableImportance.TreeVotes[] |
VariableImportance.collectVotes(int trees,
int nclasses,
Frame f,
int ncols,
float rate,
int variable,
SpeeDRFModel model,
Vec resp) |
static String[] |
CMTask.domain(int N,
Vec modelCol,
Vec dataCol,
int[] modelEnumMapping,
int[] dataEnumMapping) |
String[] |
CMTask.domain(Vec modelResp)
Compute confusion matrix domain based on model and data key.
|
static String[] |
CMTask.domain(Vec modelCol,
Vec dataCol) |
protected VarImp |
SpeeDRFModel.doVarImpCalc(Frame fr,
SpeeDRFModel model,
Vec resp) |
static CMTask |
CMTask.scoreTask(SpeeDRFModel model,
int treesToUse,
boolean computeOOB,
Frame fr,
Vec resp) |
| Modifier and Type | Field and Description |
|---|---|
protected Vec[] |
Job.ValidatedJob._train |
protected Vec[] |
Job.ValidatedJob._valid |
protected Vec |
Job.ValidatedJob._validResponse
Validation vector extracted from validation frame.
|
Vec |
Job.ColumnsResJob.response |
| Modifier and Type | Method and Description |
|---|---|
Vec |
Job.ValidatedJob.Response2CMAdaptor.adaptModelResponse2CM(Vec v)
Adapt given vector produced by a model to confusion matrix domain.
|
Vec |
Job.ValidatedJob.Response2CMAdaptor.adaptValidResponse2CM(Vec v)
Adapt given validation vector to confusion matrix domain.
|
Vec |
Job.ValidatedJob.Response2CMAdaptor.getAdaptedValidationResponse2CM()
Return cached validation response already adapted to CM domain.
|
protected Vec[] |
Job.ColumnsJob.selectVecs(Frame frame) |
Vec |
MRTask2.vecs(int i)
Returns a Vec from the Frame.
|
| Modifier and Type | Method and Description |
|---|---|
Vec |
Job.ValidatedJob.Response2CMAdaptor.adaptModelResponse2CM(Vec v)
Adapt given vector produced by a model to confusion matrix domain.
|
Vec |
Job.ValidatedJob.Response2CMAdaptor.adaptValidResponse2CM(Vec v)
Adapt given validation vector to confusion matrix domain.
|
void |
MRTask2.asyncExec(Vec... vecs) |
double |
Model.calcError(Frame ftest,
Vec vactual,
Frame fpreds,
Frame hitratio_fpreds,
String label,
boolean printMe,
int max_conf_mat_size,
ConfusionMatrix cm,
AUC auc,
HitRatio hr)
Compute the model error for a given test data set
For multi-class classification, this is the classification error based on assigning labels for the highest predicted per-class probability.
|
T |
MRTask2.dfork(int outputs,
Vec... vecs) |
T |
MRTask2.dfork(Vec... vecs)
Invokes the map/reduce computation over the given Frame.
|
T |
MRTask2.doAll(int outputs,
Vec... vecs) |
T |
MRTask2.doAll(Vec... vecs)
Invokes the map/reduce computation over the given Vecs.
|
void |
MRTask2.exec(Vec... vecs) |
protected String[] |
Job.ValidatedJob.getVectorDomain(Vec v) |
protected void |
Func.gtrash(Vec... vec)
Append given vector to clean up list.
|
protected void |
Func.ltrash(Vec... vec)
Put given vectors into local trash.
|
void |
Model.scoreCrossValidation(Job.ValidatedJob job,
Frame source,
Vec response,
Frame[] cv_preds,
long[] offsets)
Compute the cross validation error from an array of predictions for N folds.
|
| Modifier and Type | Field and Description |
|---|---|
Vec |
QuantilesPage.column |
Vec |
Impute.column |
Vec |
HitRatio.vactual |
Vec |
GainsLiftTable.vactual |
Vec |
ConfusionMatrix.vactual |
Vec |
AUC.vactual |
Vec |
GainsLiftTable.vpredict |
Vec |
ConfusionMatrix.vpredict |
Vec |
AUC.vpredict |
| Modifier and Type | Method and Description |
|---|---|
protected Vec |
RequestArguments.FrameKeyVec.defaultValue() |
protected Vec |
RequestArguments.FrameClassVec.defaultValue() |
protected Vec |
RequestArguments.FrameKeyVec.parse(String input) |
| Modifier and Type | Method and Description |
|---|---|
void |
RequestArguments.FrameKeyMultiVec.checkLegality(Vec v) |
static String |
Inspect2.x0(Vec v,
long row) |
static String |
Inspect2.x1(Vec v,
long row,
double d) |
| Modifier and Type | Method and Description |
|---|---|
void |
Env.popVec(Vec vec) |
void |
Env.subRef(Vec vec) |
Futures |
Env.subRef(Vec vec,
Futures fs)
Subtract reference count.
|
| Modifier and Type | Class and Description |
|---|---|
class |
AppendableVec
A NEW single distributed vector column.
|
class |
ByteVec
A vector of plain Bytes.
|
class |
FileVec |
class |
HdfsFileVec |
class |
NFSFileVec |
class |
S3FileVec |
class |
SubsetVec
A simple wrapper for looking at only a subset of rows
|
class |
TachyonFileVec |
class |
TransfVec
Dummy vector transforming values of given vector according to given domain mapping.
|
class |
UploadFileVec
Build a Vec by reading from an InputStream
|
class |
WrappedVec
A simple wrapper over another Vec.
|
| Modifier and Type | Field and Description |
|---|---|
Vec |
Chunk._vec |
| Modifier and Type | Method and Description |
|---|---|
Vec |
Vec.align(Vec vec)
Always makes a copy of the given vector which shares the same
group.
|
Vec |
Frame.anyVec()
Returns the first readable vector.
|
Vec |
AppendableVec.close(Futures fs) |
Vec |
TransfVec.compose(int[][] transfMap,
String[] domain)
Compose this vector with given transformation.
|
static Vec |
TransfVec.compose(TransfVec origVec,
int[][] transfMap,
String[] domain,
boolean keepOrig)
Compose given origVector with given transformation.
|
Vec |
Frame.factor(int col) |
Vec |
Frame.lastVec() |
static Vec |
Vec.make1Elem(double d)
Create a new 1-element vector in the shared vector group for 1-element vectors.
|
static Vec |
Vec.make1Elem(Key key,
double d)
Create a new 1-element vector representing a scalar value.
|
Vec |
Vec.makeCon(double d) |
Vec |
Vec.makeCon(long l)
Make a new vector with the same size and data layout as the old one, and
initialized to a constant.
|
Vec |
Vec.makeCon(long l,
String[] domain) |
Vec[] |
Vec.makeCons(int n,
long l,
String[][] domain,
boolean[] uuids,
byte[] times) |
static Vec |
Vec.makeConSeq(double x,
long len) |
static Vec[] |
Vec.makeNewCons(long rows,
int cols,
long val,
String[][] domain)
Create an array of Vecs from scratch
|
static Vec |
Vec.makeSeq(long len) |
Vec |
Vec.makeTransf(int[][] map,
String[] finalDomain)
Create a vector transforming values according given domain map.
|
Vec |
Vec.makeTransf(int[] values,
int[] indexes,
String[] domain)
Creates a new transformation from given values to given indexes of
given domain.
|
Vec |
Vec.makeZero()
Make a new vector with the same size and data layout as the old one, and
initialized to zero.
|
Vec |
Vec.makeZero(String[] domain) |
Vec[] |
Vec.makeZeros(int n) |
Vec[] |
Vec.makeZeros(int n,
String[][] domain,
boolean[] uuids,
byte[] times) |
Vec |
WrappedVec.masterVec() |
Vec |
Vec.masterVec()
This Vec does not have dependent hidden Vec it uses.
|
Vec[] |
Frame.reloadVecs() |
Vec |
Frame.remove(int idx)
Removes a numbered column.
|
Vec[] |
Frame.remove(int[] idxs)
Removes a numbered column.
|
Vec[] |
Frame.remove(int startIdx,
int endIdx)
Remove given interval of columns from frame.
|
Vec |
Frame.remove(String name)
Removes the first column with a matching name.
|
Vec |
Frame.replace(int col,
Vec nv) |
Vec |
Vec.rollupStats()
Compute the roll-up stats as-needed, and copy into the Vec object
|
Vec |
Vec.rollupStats(Futures fs) |
Vec |
SubsetVec.rows() |
Vec |
Frame.theVec(String err) |
Vec |
Vec.toEnum()
Transform this vector to enum.
|
Vec |
Frame.vec(int idx)
Returns the vector by given index.
|
Vec |
Frame.vec(String name) |
Vec[] |
Frame.vecs() |
Vec[] |
Frame.vecs(int[] idxs) |
| Modifier and Type | Method and Description |
|---|---|
Frame |
Frame.add(String name,
Vec vec)
Appends a named column, keeping the last Vec as the response
|
Vec |
Vec.align(Vec vec)
Always makes a copy of the given vector which shares the same
group.
|
int |
Frame.find(Vec vec) |
Frame |
Frame.prepend(String name,
Vec vec)
Insert a named column as the first column
|
Vec |
Frame.replace(int col,
Vec nv) |
| Constructor and Description |
|---|
Frame(Key key,
String[] names,
Vec[] vecs) |
Frame(String[] names,
Vec[] vecs) |
Frame(Vec... vecs) |
NewChunk(Vec vec,
int cidx) |
NewChunk(Vec vec,
int cidx,
int len) |
RebalanceDataSet.RebalanceTask(H2O.H2OCountedCompleter cmp,
Vec... srcVecs) |
Vec.CollectDomain(Vec v) |
Vec(Key key,
Vec v) |
| Modifier and Type | Method and Description |
|---|---|
static Vec |
MRUtils.getL2(Frame fr,
double[] scale)
Compute the L2 norm for each row of the frame
|
| Modifier and Type | Method and Description |
|---|---|
static double[] |
Utils.asDoubles(Vec v) |
static int[] |
Utils.asInts(Vec v) |
static Frame |
FrameUtils.frame(String[] names,
Vec[] vecs)
Create a new frame based on given column data.
|
static Frame |
FrameUtils.frame(String name,
Vec vec)
Create a frame with single column represented by given vector.
|
static Frame |
MRUtils.sampleFrameStratified(Frame fr,
Vec label,
float[] sampling_ratios,
long seed,
boolean debug)
Stratified sampling
|
static Frame |
MRUtils.sampleFrameStratified(Frame fr,
Vec label,
float[] sampling_ratios,
long maxrows,
long seed,
boolean allowOversampling,
boolean verbose)
Stratified sampling for classifiers
|
| Constructor and Description |
|---|
MRUtils.ClassDist(Vec label) |
Copyright © 2013–2014 0xdata. All rights reserved.