Class CHashedMultilabelModel represents application specific model and contains application dependent logic for solving multilabel classification with feature hashing within a generic SO framework. We hash the feature of each class with a separate seed and put them in the same feature space (exploded feature space).
The hashing function is defined as per:
[1] K. Weinberger, A. Dasgupta, J. Attenberg, J. Langford, A. Smola Feature Hashing for Large Scale Multitask Learning http://alex.smola.org/papers/2009/Weinbergeretal09.pdf
在文件 HashedMultilabelModel.h 第 28 行定义.
Public 成员函数 | |
CHashedMultilabelModel () | |
CHashedMultilabelModel (CFeatures *features, CStructuredLabels *labels, int32_t dim) | |
virtual | ~CHashedMultilabelModel () |
virtual CStructuredLabels * | structured_labels_factory (int32_t num_examples=0) |
virtual int32_t | get_dim () const |
virtual SGVector< float64_t > | get_joint_feature_vector (int32_t feat_idx, CStructuredData *y) |
virtual SGSparseVector< float64_t > | get_sparse_joint_feature_vector (int32_t feat_idx, CStructuredData *y) |
virtual CResultSet * | argmax (SGVector< float64_t > w, int32_t feat_idx, bool const training=true) |
virtual float64_t | delta_loss (CStructuredData *y1, CStructuredData *y2) |
virtual void | set_misclass_cost (float64_t false_positive, float64_t false_negative) |
virtual void | init_primal_opt (float64_t regularization, SGMatrix< float64_t > &A, SGVector< float64_t > a, SGMatrix< float64_t > B, SGVector< float64_t > &b, SGVector< float64_t > &lb, SGVector< float64_t > &ub, SGMatrix< float64_t > &C) |
virtual void | set_seeds (SGVector< uint32_t > seeds) |
virtual const char * | get_name () const |
void | set_labels (CStructuredLabels *labs) |
CStructuredLabels * | get_labels () |
void | set_features (CFeatures *feats) |
CFeatures * | get_features () |
SGVector< float64_t > | get_joint_feature_vector (int32_t feat_idx, int32_t lab_idx) |
SGSparseVector< float64_t > | get_sparse_joint_feature_vector (int32_t feat_idx, int32_t lab_idx) |
float64_t | delta_loss (int32_t ytrue_idx, CStructuredData *ypred) |
virtual void | init_training () |
virtual bool | check_training_setup () const |
virtual int32_t | get_num_aux () const |
virtual int32_t | get_num_aux_con () const |
virtual CSGObject * | shallow_copy () const |
virtual CSGObject * | deep_copy () const |
virtual bool | is_generic (EPrimitiveType *generic) const |
template<class T > | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
template<> | |
void | set_generic () |
void | unset_generic () |
virtual void | print_serializable (const char *prefix="") |
virtual bool | save_serializable (CSerializableFile *file, const char *prefix="") |
virtual bool | load_serializable (CSerializableFile *file, const char *prefix="") |
void | set_global_io (SGIO *io) |
SGIO * | get_global_io () |
void | set_global_parallel (Parallel *parallel) |
Parallel * | get_global_parallel () |
void | set_global_version (Version *version) |
Version * | get_global_version () |
SGStringList< char > | get_modelsel_names () |
void | print_modsel_params () |
char * | get_modsel_param_descr (const char *param_name) |
index_t | get_modsel_param_index (const char *param_name) |
void | build_gradient_parameter_dictionary (CMap< TParameter *, CSGObject * > *dict) |
virtual void | update_parameter_hash () |
virtual bool | parameter_hash_changed () |
virtual bool | equals (CSGObject *other, float64_t accuracy=0.0, bool tolerant=false) |
virtual CSGObject * | clone () |
Public 属性 | |
SGIO * | io |
Parallel * | parallel |
Version * | version |
Parameter * | m_parameters |
Parameter * | m_model_selection_parameters |
Parameter * | m_gradient_parameters |
uint32_t | m_hash |
Protected 成员函数 | |
virtual void | load_serializable_pre () throw (ShogunException) |
virtual void | load_serializable_post () throw (ShogunException) |
virtual void | save_serializable_pre () throw (ShogunException) |
virtual void | save_serializable_post () throw (ShogunException) |
Protected 属性 | |
CStructuredLabels * | m_labels |
CFeatures * | m_features |
default constructor
在文件 HashedMultilabelModel.cpp 第 18 行定义.
CHashedMultilabelModel | ( | CFeatures * | features, |
CStructuredLabels * | labels, | ||
int32_t | dim | ||
) |
constructor
features | features |
labels | structured labels |
dim | new joint feature space dimension |
在文件 HashedMultilabelModel.cpp 第 24 行定义.
|
virtual |
destructor
在文件 HashedMultilabelModel.cpp 第 30 行定义.
|
virtual |
obtain the argmax of
\[ \Delta(y_{pred}, y_{truth} + \langle w, \Psi(x_{truth}, y_{pred}) \rangle \]
w | weight vector |
feat_idx | index of the feature to compute the argmax |
training | true if argmax is called during training Then, it is assumed that the label indexed by feat_idx in m_labels corresponds to the true label of the corresponding feature vector |
实现了 CStructuredModel.
在文件 HashedMultilabelModel.cpp 第 236 行定义.
|
inherited |
Builds a dictionary of all parameters in SGObject as well of those of SGObjects that are parameters of this object. Dictionary maps parameters to the objects that own them.
dict | dictionary of parameters to be built. |
在文件 SGObject.cpp 第 597 行定义.
|
virtualinherited |
method to be called from a SO machine before training to ensure that the training data is valid (e.g. check that there is at least one example for every class). In this class the method is empty and it can be re-implemented for any application (e.g. HM-SVM).
被 CHMSVMModel 重载.
在文件 StructuredModel.cpp 第 183 行定义.
|
virtualinherited |
Creates a clone of the current object. This is done via recursively traversing all parameters, which corresponds to a deep copy. Calling equals on the cloned object always returns true although none of the memory of both objects overlaps.
在文件 SGObject.cpp 第 714 行定义.
|
virtualinherited |
A deep copy. All the instance variables will also be copied.
在文件 SGObject.cpp 第 198 行定义.
|
virtual |
compute
\[ \Delta(y_{1}, y_{2}) \]
y1 | an instance of structured data |
y2 | another instance of structured data |
重载 CStructuredModel .
在文件 HashedMultilabelModel.cpp 第 131 行定义.
|
inherited |
computes \( \Delta(y_{\text{true}}, y_{\text{pred}}) \)
ytrue_idx | index of the true label in labels |
ypred | the predicted label |
在文件 StructuredModel.cpp 第 147 行定义.
Recursively compares the current SGObject to another one. Compares all registered numerical parameters, recursion upon complex (SGObject) parameters. Does not compare pointers!
May be overwritten but please do with care! Should not be necessary in most cases.
other | object to compare with |
accuracy | accuracy to use for comparison (optional) |
tolerant | allows linient check on float equality (within accuracy) |
在文件 SGObject.cpp 第 618 行定义.
|
virtual |
实现了 CStructuredModel.
在文件 HashedMultilabelModel.cpp 第 69 行定义.
|
inherited |
|
inherited |
|
inherited |
|
inherited |
|
virtual |
get joint feature vector
\[ \vec{\Psi}(\bf{x}_\text{feat\_idx}, \bf{y}) \]
feat_idx | index of the feature vector to use |
y | structured label to use |
重载 CStructuredModel .
在文件 HashedMultilabelModel.cpp 第 88 行定义.
gets joint feature vector
\[ \vec{\Psi}(\bf{x}_\text{feat\_idx}, \bf{y}_\text{lab\_idx}) \]
feat_idx | index of the feature vector to use |
lab_idx | index of the structured label to use |
在文件 StructuredModel.cpp 第 105 行定义.
|
inherited |
|
inherited |
在文件 SGObject.cpp 第 498 行定义.
|
inherited |
Returns description of a given parameter string, if it exists. SG_ERROR otherwise
param_name | name of the parameter |
在文件 SGObject.cpp 第 522 行定义.
|
inherited |
Returns index of model selection parameter with provided index
param_name | name of model selection parameter |
在文件 SGObject.cpp 第 535 行定义.
|
virtual |
|
virtualinherited |
get the number of auxiliary variables to introduce in the optimization problem. By default, this class do not impose the use of auxiliary variables and it will return zero. Re-implement this method subclasses to use auxiliary variables.
return the number of auxiliary variables
被 CHMSVMModel 重载.
在文件 StructuredModel.cpp 第 189 行定义.
|
virtualinherited |
get the number of auxiliary constraints to introduce in the optimization problem. By default, this class do not impose the use of any auxiliary constraints and it will return zero. Re-implement this method in subclasses to use auxiliary constraints.
return the number of auxiliary constraints
被 CHMSVMModel 重载.
在文件 StructuredModel.cpp 第 194 行定义.
|
virtual |
get joint feature vector
\[ \vec{\Psi}(\bf{x}_\text{feat\_idx}, \bf{y}) \]
feat_idx | index of the feature vector to use |
y | structured label to use |
重载 CStructuredModel .
在文件 HashedMultilabelModel.cpp 第 97 行定义.
|
inherited |
gets joint feature vector
\[ \vec{\Psi}(\bf{x}_\text{feat\_idx}, \bf{y}_\text{lab\_idx}) \]
feat_idx | index of the feature vector to use |
lab_idx | index of the structured label to use |
在文件 StructuredModel.cpp 第 126 行定义.
|
virtual |
intialize the optimization problem
regularization | regularization strength |
A | is [-dPsi(y) | -I_N ] with M+N columns => max, M+1 nnz per row |
a | unused input |
B | unused input |
b | upper bounds of the constraint, Ax <= b |
lb | lower bounds for w |
ub | upper bounds for w |
C | regularization matrix, w'Cw |
重载 CStructuredModel .
在文件 HashedMultilabelModel.cpp 第 168 行定义.
|
virtualinherited |
initializes the part of the model that needs to be used during training. In this class this method is empty and it can be re-implemented for any particular StructuredModel
被 CHMSVMModel , 以及 CFactorGraphModel 重载.
在文件 StructuredModel.cpp 第 178 行定义.
|
virtualinherited |
If the SGSerializable is a class template then TRUE will be returned and GENERIC is set to the type of the generic.
generic | set to the type of the generic if returning TRUE |
在文件 SGObject.cpp 第 296 行定义.
|
virtualinherited |
Load this object from file. If it will fail (returning FALSE) then this object will contain inconsistent data and should not be used!
file | where to load from |
prefix | prefix for members |
在文件 SGObject.cpp 第 369 行定义.
|
protectedvirtualinherited |
Can (optionally) be overridden to post-initialize some member variables which are not PARAMETER::ADD'ed. Make sure that at first the overridden method BASE_CLASS::LOAD_SERIALIZABLE_POST is called.
ShogunException | will be thrown if an error occurs. |
被 CKernel, CWeightedDegreePositionStringKernel, CList, CAlphabet, CLinearHMM, CGaussianKernel, CInverseMultiQuadricKernel, CCircularKernel , 以及 CExponentialKernel 重载.
在文件 SGObject.cpp 第 426 行定义.
|
protectedvirtualinherited |
Can (optionally) be overridden to pre-initialize some member variables which are not PARAMETER::ADD'ed. Make sure that at first the overridden method BASE_CLASS::LOAD_SERIALIZABLE_PRE is called.
ShogunException | will be thrown if an error occurs. |
被 CDynamicArray< T >, CDynamicArray< float64_t >, CDynamicArray< float32_t >, CDynamicArray< int32_t >, CDynamicArray< char >, CDynamicArray< bool > , 以及 CDynamicObjectArray 重载.
在文件 SGObject.cpp 第 421 行定义.
|
virtualinherited |
在文件 SGObject.cpp 第 262 行定义.
|
inherited |
prints all parameter registered for model selection and their type
在文件 SGObject.cpp 第 474 行定义.
|
virtualinherited |
|
virtualinherited |
Save this object to file.
file | where to save the object; will be closed during returning if PREFIX is an empty string. |
prefix | prefix for members |
在文件 SGObject.cpp 第 314 行定义.
|
protectedvirtualinherited |
Can (optionally) be overridden to post-initialize some member variables which are not PARAMETER::ADD'ed. Make sure that at first the overridden method BASE_CLASS::SAVE_SERIALIZABLE_POST is called.
ShogunException | will be thrown if an error occurs. |
被 CKernel 重载.
在文件 SGObject.cpp 第 436 行定义.
|
protectedvirtualinherited |
Can (optionally) be overridden to pre-initialize some member variables which are not PARAMETER::ADD'ed. Make sure that at first the overridden method BASE_CLASS::SAVE_SERIALIZABLE_PRE is called.
ShogunException | will be thrown if an error occurs. |
被 CKernel, CDynamicArray< T >, CDynamicArray< float64_t >, CDynamicArray< float32_t >, CDynamicArray< int32_t >, CDynamicArray< char >, CDynamicArray< bool > , 以及 CDynamicObjectArray 重载.
在文件 SGObject.cpp 第 431 行定义.
|
inherited |
|
inherited |
在文件 SGObject.cpp 第 41 行定义.
|
inherited |
在文件 SGObject.cpp 第 46 行定义.
|
inherited |
在文件 SGObject.cpp 第 51 行定义.
|
inherited |
在文件 SGObject.cpp 第 56 行定义.
|
inherited |
在文件 SGObject.cpp 第 61 行定义.
|
inherited |
在文件 SGObject.cpp 第 66 行定义.
|
inherited |
在文件 SGObject.cpp 第 71 行定义.
|
inherited |
在文件 SGObject.cpp 第 76 行定义.
|
inherited |
在文件 SGObject.cpp 第 81 行定义.
|
inherited |
在文件 SGObject.cpp 第 86 行定义.
|
inherited |
在文件 SGObject.cpp 第 91 行定义.
|
inherited |
在文件 SGObject.cpp 第 96 行定义.
|
inherited |
在文件 SGObject.cpp 第 101 行定义.
|
inherited |
在文件 SGObject.cpp 第 106 行定义.
|
inherited |
在文件 SGObject.cpp 第 111 行定义.
|
inherited |
set generic type to T
|
inherited |
|
inherited |
|
inherited |
|
inherited |
set misclassification cost for false positive and false negative
false_positive | cost for false positive |
false_negative | cost for false negative |
在文件 HashedMultilabelModel.cpp 第 74 行定义.
|
virtual |
set seeds used for hashing features of each class
seeds | vector of seeds used for hashing |
在文件 HashedMultilabelModel.cpp 第 81 行定义.
|
virtualinherited |
A shallow copy. All the SGObject instance variables will be simply assigned and SG_REF-ed.
被 CGaussianKernel 重载.
在文件 SGObject.cpp 第 192 行定义.
|
virtual |
|
inherited |
unset generic type
this has to be called in classes specializing a template class
在文件 SGObject.cpp 第 303 行定义.
|
virtualinherited |
Updates the hash of current parameter combination
在文件 SGObject.cpp 第 248 行定义.
|
inherited |
io
在文件 SGObject.h 第 369 行定义.
|
protectedinherited |
feature vectors
在文件 StructuredModel.h 第 308 行定义.
|
inherited |
parameters wrt which we can compute gradients
在文件 SGObject.h 第 384 行定义.
|
inherited |
Hash of parameter values
在文件 SGObject.h 第 387 行定义.
|
protectedinherited |
structured labels
在文件 StructuredModel.h 第 305 行定义.
|
inherited |
model selection parameters
在文件 SGObject.h 第 381 行定义.
|
inherited |
parameters
在文件 SGObject.h 第 378 行定义.
|
inherited |
parallel
在文件 SGObject.h 第 372 行定义.
|
inherited |
version
在文件 SGObject.h 第 375 行定义.