Hendrik/xgboost
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

add pathdump

Browse Source
This commit is contained in:
tqchen
2014-02-26 17:08:23 -08:00
parent 88c982012a
commit 7b2fe1bf5d
7 changed files with 91 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

42
booster/tree/xgboost_tree.hpp
View File

@@ -57,7 +57,7 @@ namespace xgboost{
virtual int GetLeafIndex( const std::vector<float> &feat,
const std::vector<bool> &funknown,
unsigned gid = 0 ){
unsigned gid = 0 ){
// start from groups that belongs to current data
int pid = (int)gid;
// tranverse tree
@@ -67,18 +67,28 @@ namespace xgboost{
}
return pid;
}
virtual void PredPath( std::vector<int> &path, const FMatrixS::Line &feat, unsigned gid = 0 ){
path.clear();
this->InitTmp();
this->PrepareTmp( feat );
int pid = (int)gid;
path.push_back( pid );
// tranverse tree
while( !tree[ pid ].is_leaf() ){
unsigned split_index = tree[ pid ].split_index();
pid = this->GetNext( pid, tmp_feat[ split_index ], tmp_funknown[ split_index ] );
path.push_back( pid );
}
this->DropTmp( feat );
}
virtual float Predict( const FMatrixS::Line &feat, unsigned gid = 0 ){
this->InitTmp();
for( unsigned i = 0; i < feat.len; i ++ ){
utils::Assert( feat[i].findex < (unsigned)tmp_funknown.size() , "input feature execeed bound" );
tmp_funknown[ feat[i].findex ] = false;
tmp_feat[ feat[i].findex ] = feat[i].fvalue;
}
this->PrepareTmp( feat );
int pid = this->GetLeafIndex( tmp_feat, tmp_funknown, gid );
// set back
for( unsigned i = 0; i < feat.len; i ++ ){
tmp_funknown[ feat[i].findex ] = true;
}
this->DropTmp( feat );
return tree[ pid ].leaf_value();
}
virtual float Predict( const std::vector<float> &feat,
@@ -127,6 +137,18 @@ namespace xgboost{
std::fill( tmp_funknown.begin(), tmp_funknown.end(), true );
}
}
inline void PrepareTmp( const FMatrixS::Line &feat ){
for( unsigned i = 0; i < feat.len; i ++ ){
utils::Assert( feat[i].findex < (unsigned)tmp_funknown.size() , "input feature execeed bound" );
tmp_funknown[ feat[i].findex ] = false;
tmp_feat[ feat[i].findex ] = feat[i].fvalue;
}
}
inline void DropTmp( const FMatrixS::Line &feat ){
for( unsigned i = 0; i < feat.len; i ++ ){
tmp_funknown[ feat[i].findex ] = true;
}
}
inline int GetNext( int pid, float fvalue, bool is_unknown ){
float split_value = tree[ pid ].split_cond();

8
booster/xgboost.h
View File

@@ -65,6 +65,14 @@ namespace xgboost{
std::vector<float> &hess,
const FMatrixS &feats,
const std::vector<unsigned> &root_index ) = 0;
/*!
* \brief predict the path ids along a trees, for given sparse feature vector. When booster is a tree
* \param path the result of path
* \param rid root id of current instance, default = 0
*/
virtual void PredPath( std::vector<int> &path, const FMatrixS::Line &feat, unsigned rid = 0 ){
utils::Error( "not implemented" );
}
/*!
* \brief predict values for given sparse feature vector
* NOTE: in tree implementation, this is not threadsafe, used dense version to ensure threadsafety

23
booster/xgboost_gbmbase.h
View File

@@ -179,6 +179,25 @@ namespace xgboost{
boosters[i]->DumpModel( fo );
}
}
/*!
* \brief Dump path of all trees
* \param fo text file
* \param data input data
*/
inline void DumpPath( FILE *fo, const FMatrixS &data ){
for( size_t i = 0; i < data.NumRow(); ++ i ){
for( size_t j = 0; j < boosters.size(); ++ j ){
if( j != 0 ) fprintf( fo, "\t" );
std::vector<int> path;
boosters[j]->PredPath( path, data[i] );
fprintf( fo, "%d", path[0] );
for( size_t k = 1; k < path.size(); ++ k ){
fprintf( fo, ",%d", path[k] );
}
}
fprintf( fo, "\n" );
}
}
public:
/*!
* \brief do gradient boost training for one step, using the information given
@@ -195,7 +214,7 @@ namespace xgboost{
const std::vector<unsigned> &root_index ) {
booster::IBooster *bst = this->GetUpdateBooster();
bst->DoBoost( grad, hess, feats, root_index );
}
}
/*!
* \brief predict values for given sparse feature vector
* NOTE: in tree implementation, this is not threadsafe
@@ -204,7 +223,7 @@ namespace xgboost{
* \param rid root id of current instance, default = 0
* \return prediction
*/
virtual float Predict( const booster::FMatrixS::Line &feat, int buffer_index = -1, unsigned rid = 0 ){
virtual float Predict( const booster::FMatrixS::Line &feat, int buffer_index = -1, unsigned rid = 0 ){
size_t istart = 0;
float psum = 0.0f;