xgboost
Public Member Functions | Static Public Member Functions | Protected Attributes | List of all members
xgboost::Predictor Class Referenceabstract

Performs prediction on individual training instances or batches of instances for GBTree. Prediction functions all take a GBTreeModel and a DMatrix as input and output a vector of predictions. The predictor does not modify any state of the model itself. More...

#include <predictor.h>

Collaboration diagram for xgboost::Predictor:
Collaboration graph

Public Member Functions

 Predictor (GenericParameter const *ctx)
 
virtual ~Predictor ()=default
 
virtual void Configure (const std::vector< std::pair< std::string, std::string >> &)
 Configure and register input matrices in prediction cache. More...
 
void InitOutPredictions (const MetaInfo &info, HostDeviceVector< bst_float > *out_predt, const gbm::GBTreeModel &model) const
 Initialize output prediction. More...
 
virtual void PredictBatch (DMatrix *dmat, PredictionCacheEntry *out_preds, const gbm::GBTreeModel &model, uint32_t tree_begin, uint32_t tree_end=0) const =0
 Generate batch predictions for a given feature matrix. May use cached predictions if available instead of calculating from scratch. More...
 
virtual bool InplacePredict (dmlc::any const &x, std::shared_ptr< DMatrix > p_m, const gbm::GBTreeModel &model, float missing, PredictionCacheEntry *out_preds, uint32_t tree_begin=0, uint32_t tree_end=0) const =0
 Inplace prediction. More...
 
virtual void PredictInstance (const SparsePage::Inst &inst, std::vector< bst_float > *out_preds, const gbm::GBTreeModel &model, unsigned tree_end=0) const =0
 online prediction function, predict score for one instance at a time NOTE: use the batch prediction interface if possible, batch prediction is usually more efficient than online prediction This function is NOT threadsafe, make sure you only call from one thread. More...
 
virtual void PredictLeaf (DMatrix *dmat, HostDeviceVector< bst_float > *out_preds, const gbm::GBTreeModel &model, unsigned tree_end=0) const =0
 predict the leaf index of each tree, the output will be nsample * ntree vector this is only valid in gbtree predictor. More...
 
virtual void PredictContribution (DMatrix *dmat, HostDeviceVector< bst_float > *out_contribs, const gbm::GBTreeModel &model, unsigned tree_end=0, std::vector< bst_float > const *tree_weights=nullptr, bool approximate=false, int condition=0, unsigned condition_feature=0) const =0
 feature contributions to individual predictions; the output will be a vector of length (nfeats + 1) * num_output_group * nsample, arranged in that order. More...
 
virtual void PredictInteractionContributions (DMatrix *dmat, HostDeviceVector< bst_float > *out_contribs, const gbm::GBTreeModel &model, unsigned tree_end=0, std::vector< bst_float > const *tree_weights=nullptr, bool approximate=false) const =0
 

Static Public Member Functions

static PredictorCreate (std::string const &name, GenericParameter const *generic_param)
 Creates a new Predictor*. More...
 

Protected Attributes

GenericParameter const * ctx_
 

Detailed Description

Performs prediction on individual training instances or batches of instances for GBTree. Prediction functions all take a GBTreeModel and a DMatrix as input and output a vector of predictions. The predictor does not modify any state of the model itself.

Constructor & Destructor Documentation

◆ Predictor()

xgboost::Predictor::Predictor ( GenericParameter const *  ctx)
inlineexplicit

◆ ~Predictor()

virtual xgboost::Predictor::~Predictor ( )
virtualdefault

Member Function Documentation

◆ Configure()

virtual void xgboost::Predictor::Configure ( const std::vector< std::pair< std::string, std::string >> &  )
virtual

Configure and register input matrices in prediction cache.

Parameters
cfgThe configuration.

◆ Create()

static Predictor* xgboost::Predictor::Create ( std::string const &  name,
GenericParameter const *  generic_param 
)
static

Creates a new Predictor*.

Parameters
nameName of the predictor.
generic_paramPointer to runtime parameters.

◆ InitOutPredictions()

void xgboost::Predictor::InitOutPredictions ( const MetaInfo info,
HostDeviceVector< bst_float > *  out_predt,
const gbm::GBTreeModel &  model 
) const

Initialize output prediction.

Parameters
infoMeta info for the DMatrix object used for prediction.
out_predtPrediction vector to be initialized.
modelTree model used for prediction.

◆ InplacePredict()

virtual bool xgboost::Predictor::InplacePredict ( dmlc::any const &  x,
std::shared_ptr< DMatrix p_m,
const gbm::GBTreeModel &  model,
float  missing,
PredictionCacheEntry out_preds,
uint32_t  tree_begin = 0,
uint32_t  tree_end = 0 
) const
pure virtual

Inplace prediction.

Parameters
xType erased data adapter.
modelThe model to predict from.
missingMissing value in the data.
[in,out]out_predsThe output preds.
tree_begin(Optional) Beginning of boosted trees used for prediction.
tree_end(Optional) End of booster trees. 0 means do not limit trees.
Returns
True if the data can be handled by current predictor, false otherwise.

◆ PredictBatch()

virtual void xgboost::Predictor::PredictBatch ( DMatrix dmat,
PredictionCacheEntry out_preds,
const gbm::GBTreeModel &  model,
uint32_t  tree_begin,
uint32_t  tree_end = 0 
) const
pure virtual

Generate batch predictions for a given feature matrix. May use cached predictions if available instead of calculating from scratch.

Parameters
[in,out]dmatFeature matrix.
[in,out]out_predsThe output preds.
modelThe model to predict from.
tree_beginThe tree begin index.
tree_endThe tree end index.

◆ PredictContribution()

virtual void xgboost::Predictor::PredictContribution ( DMatrix dmat,
HostDeviceVector< bst_float > *  out_contribs,
const gbm::GBTreeModel &  model,
unsigned  tree_end = 0,
std::vector< bst_float > const *  tree_weights = nullptr,
bool  approximate = false,
int  condition = 0,
unsigned  condition_feature = 0 
) const
pure virtual

feature contributions to individual predictions; the output will be a vector of length (nfeats + 1) * num_output_group * nsample, arranged in that order.

Parameters
[in,out]dmatThe input feature matrix.
[in,out]out_contribsThe output feature contribs.
modelModel to make predictions from.
tree_endThe tree end index.
tree_weights(Optional) Weights to multiply each tree by.
approximateUse fast approximate algorithm.
conditionCondition on the condition_feature (0=no, -1=cond off, 1=cond on).
condition_featureFeature to condition on (i.e. fix) during calculations.

◆ PredictInstance()

virtual void xgboost::Predictor::PredictInstance ( const SparsePage::Inst inst,
std::vector< bst_float > *  out_preds,
const gbm::GBTreeModel &  model,
unsigned  tree_end = 0 
) const
pure virtual

online prediction function, predict score for one instance at a time NOTE: use the batch prediction interface if possible, batch prediction is usually more efficient than online prediction This function is NOT threadsafe, make sure you only call from one thread.

Parameters
instThe instance to predict.
[in,out]out_predsThe output preds.
modelThe model to predict from
tree_end(Optional) The tree end index.

◆ PredictInteractionContributions()

virtual void xgboost::Predictor::PredictInteractionContributions ( DMatrix dmat,
HostDeviceVector< bst_float > *  out_contribs,
const gbm::GBTreeModel &  model,
unsigned  tree_end = 0,
std::vector< bst_float > const *  tree_weights = nullptr,
bool  approximate = false 
) const
pure virtual

◆ PredictLeaf()

virtual void xgboost::Predictor::PredictLeaf ( DMatrix dmat,
HostDeviceVector< bst_float > *  out_preds,
const gbm::GBTreeModel &  model,
unsigned  tree_end = 0 
) const
pure virtual

predict the leaf index of each tree, the output will be nsample * ntree vector this is only valid in gbtree predictor.

Parameters
[in,out]dmatThe input feature matrix.
[in,out]out_predsThe output preds.
modelModel to make predictions from.
tree_end(Optional) The tree end index.

Member Data Documentation

◆ ctx_

GenericParameter const* xgboost::Predictor::ctx_
protected

The documentation for this class was generated from the following file: