PLearn 0.1
|
#include <plearn/base/Object.h>
#include <plearn/base/PP.h>
#include <plearn/math/TMat.h>
#include <plearn/math/StatsCollector.h>
#include "VMField.h"
#include <plearn/dict/Dictionary.h>
#include <plearn/io/PPath.h>
#include <map>
Go to the source code of this file.
Classes | |
class | PLearn::VMatrix |
Base classes for virtual matrices. More... | |
class | PLearn::DiffTemplate< ObjectType, VMatrix > |
class | PLearn::TypeTraits< VMatrix > |
Namespaces | |
namespace | PLearn |
< for swap | |
Defines | |
#define | SPECIAL_FORMAT ((real)3.1e36) |
NOTE: How to handle exotic cases of data-sets whose input or target are not standard Vecs: The idea is to still have the getExample build and return Vecs, but the representation of these Vecs will have a special format, detected and understood by a specialised Learner or specialised Variables: Hack: format des Vec compris par un Learner: | |
Functions | |
Object * | PLearn::toObjectPtr (const VMatrix &o) |
PStream & | PLearn::operator>> (PStream &in, VMatrix &o) |
PStream & | PLearn::operator>> (PStream &in, VMatrix *&o) |
PStream & | PLearn::operator<< (PStream &out, const VMatrix &o) |
PStream & | PLearn::operator>> (PStream &in, PP< VMatrix > &o) |
template<class ObjectType > | |
int | PLearn::diff (const string &refer, const string &other, const Option< ObjectType, VMatrix > *opt, PLearnDiff *diffs) |
Definition in file VMatrix.h.
#define SPECIAL_FORMAT ((real)3.1e36) |
NOTE: How to handle exotic cases of data-sets whose input or target are not standard Vecs: The idea is to still have the getExample build and return Vecs, but the representation of these Vecs will have a special format, detected and understood by a specialised Learner or specialised Variables: Hack: format des Vec compris par un Learner:
Si v[0] == SPECIAL_FORMAT v[1] indique le format de ce qui suit (v[2] ...): 0 sparse vec de la forme: length nvals i val i val ... 1 pointeur vers un Object de la forme: ptr (cast du float ou du double) 2 tenseur plein de la forme: rank size_1...size_n val ... 3 tenseur sparse de la forme: rank size_1...size_n nvals i_1...i_n val i1...i_n val ...