SHOGUN
v1.1.0
|
Features that compute the Weighted Degreee Kernel feature space explicitly.
Definition at line 28 of file HashedWDFeaturesTransposed.h.
Public Member Functions | |
CHashedWDFeaturesTransposed () | |
CHashedWDFeaturesTransposed (CStringFeatures< uint8_t > *str, int32_t start_order, int32_t order, int32_t from_order, int32_t hash_bits=12) | |
CHashedWDFeaturesTransposed (const CHashedWDFeaturesTransposed &orig) | |
virtual | ~CHashedWDFeaturesTransposed () |
virtual int32_t | get_dim_feature_space () const |
virtual float64_t | dot (int32_t vec_idx1, CDotFeatures *df, int32_t vec_idx2) |
virtual float64_t | dense_dot (int32_t vec_idx1, const float64_t *vec2, int32_t vec2_len) |
virtual void | dense_dot_range (float64_t *output, int32_t start, int32_t stop, float64_t *alphas, float64_t *vec, int32_t dim, float64_t b) |
virtual void | dense_dot_range_subset (int32_t *sub_index, int32_t num, float64_t *output, float64_t *alphas, float64_t *vec, int32_t dim, float64_t b) |
virtual void | add_to_dense_vec (float64_t alpha, int32_t vec_idx1, float64_t *vec2, int32_t vec2_len, bool abs_val=false) |
virtual int32_t | get_nnz_features_for_vector (int32_t num) |
virtual CFeatures * | duplicate () const |
virtual EFeatureType | get_feature_type () |
virtual EFeatureClass | get_feature_class () |
virtual int32_t | get_num_vectors () const |
virtual int32_t | get_size () |
void | set_normalization_const (float64_t n=0) |
float64_t | get_normalization_const () |
virtual void * | get_feature_iterator (int32_t vector_index) |
virtual bool | get_next_feature (int32_t &index, float64_t &value, void *iterator) |
virtual void | free_feature_iterator (void *iterator) |
virtual const char * | get_name () const |
![]() | |
CDotFeatures (int32_t size=0) | |
CDotFeatures (const CDotFeatures &orig) | |
CDotFeatures (CFile *loader) | |
virtual | ~CDotFeatures () |
float64_t | get_combined_feature_weight () |
void | set_combined_feature_weight (float64_t nw) |
SGMatrix< float64_t > | get_computed_dot_feature_matrix () |
SGVector< float64_t > | get_computed_dot_feature_vector (int32_t num) |
void | benchmark_add_to_dense_vector (int32_t repeats=5) |
void | benchmark_dense_dot_range (int32_t repeats=5) |
virtual SGVector< float64_t > | get_mean () |
virtual SGMatrix< float64_t > | get_cov () |
![]() | |
CFeatures (int32_t size=0) | |
CFeatures (const CFeatures &orig) | |
CFeatures (CFile *loader) | |
virtual | ~CFeatures () |
virtual int32_t | add_preprocessor (CPreprocessor *p) |
set preprocessor | |
virtual CPreprocessor * | del_preprocessor (int32_t num) |
del current preprocessor | |
CPreprocessor * | get_preprocessor (int32_t num) |
get current preprocessor | |
void | set_preprocessed (int32_t num) |
bool | is_preprocessed (int32_t num) |
int32_t | get_num_preprocessed () |
get whether specified preprocessor (or all if num=1) was/were already applied | |
int32_t | get_num_preprocessors () const |
void | clean_preprocessors () |
int32_t | get_cache_size () |
virtual bool | reshape (int32_t num_features, int32_t num_vectors) |
void | list_feature_obj () |
virtual void | load (CFile *loader) |
virtual void | save (CFile *writer) |
bool | check_feature_compatibility (CFeatures *f) |
bool | has_property (EFeatureProperty p) |
void | set_property (EFeatureProperty p) |
void | unset_property (EFeatureProperty p) |
virtual void | set_subset (CSubset *subset) |
virtual void | remove_subset () |
virtual void | subset_changed_post () |
index_t | subset_idx_conversion (index_t idx) const |
bool | has_subset () const |
virtual CFeatures * | copy_subset (SGVector< index_t > indices) |
![]() | |
CSGObject () | |
CSGObject (const CSGObject &orig) | |
virtual | ~CSGObject () |
virtual bool | is_generic (EPrimitiveType *generic) const |
template<class T > | |
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 () |
SGVector< char * > | get_modelsel_names () |
char * | get_modsel_param_descr (const char *param_name) |
index_t | get_modsel_param_index (const char *param_name) |
Protected Member Functions | |
void | set_wd_weights () |
![]() | |
void | display_progress (int32_t start, int32_t stop, int32_t v) |
Static Protected Member Functions | |
static void * | dense_dot_range_helper (void *p) |
Protected Attributes | |
CStringFeatures< uint8_t > * | strings |
SGString< uint8_t > * | transposed_strings |
int32_t | degree |
int32_t | start_degree |
int32_t | from_degree |
int32_t | string_length |
int32_t | num_strings |
int32_t | alphabet_size |
int32_t | w_dim |
int32_t | partial_w_dim |
float64_t * | wd_weights |
uint32_t | mask |
int32_t | m_hash_bits |
float64_t | normalization_const |
![]() | |
float64_t | combined_weight |
feature weighting in combined dot features | |
![]() | |
CSubset * | m_subset |
Additional Inherited Members | |
![]() | |
SGIO * | io |
Parallel * | parallel |
Version * | version |
Parameter * | m_parameters |
Parameter * | m_model_selection_parameters |
default constructor
Definition at line 38 of file HashedWDFeaturesTransposed.cpp.
CHashedWDFeaturesTransposed | ( | CStringFeatures< uint8_t > * | str, |
int32_t | start_order, | ||
int32_t | order, | ||
int32_t | from_order, | ||
int32_t | hash_bits = 12 |
||
) |
constructor
str | stringfeatures (of bytes) |
start_order | do degrees starting with start_order up to order |
order | of wd kernel |
from_order | use first order weights from higher order weighting |
hash_bits | number of bits in hash |
Definition at line 63 of file HashedWDFeaturesTransposed.cpp.
CHashedWDFeaturesTransposed | ( | const CHashedWDFeaturesTransposed & | orig | ) |
copy constructor
Definition at line 99 of file HashedWDFeaturesTransposed.cpp.
|
virtual |
destructor
Definition at line 115 of file HashedWDFeaturesTransposed.cpp.
|
virtual |
add vector 1 multiplied with alpha to dense vector2
alpha | scalar alpha |
vec_idx1 | index of first vector |
vec2 | pointer to real valued vector |
vec2_len | length of real valued vector |
abs_val | if true add the absolute value |
Implements CDotFeatures.
Definition at line 474 of file HashedWDFeaturesTransposed.cpp.
compute dot product between vector1 and a dense vector
vec_idx1 | index of first vector |
vec2 | pointer to real valued vector |
vec2_len | length of real valued vector |
Implements CDotFeatures.
Definition at line 157 of file HashedWDFeaturesTransposed.cpp.
|
virtual |
Compute the dot product for a range of vectors. This function makes use of dense_dot alphas[i] * sparse[i]^T * w + b
output | result for the given vector range |
start | start vector range from this idx |
stop | stop vector range at this idx |
alphas | scalars to multiply with, may be NULL |
vec | dense vector to compute dot product with |
dim | length of the dense vector |
b | bias |
Reimplemented from CDotFeatures.
Definition at line 194 of file HashedWDFeaturesTransposed.cpp.
|
staticprotected |
Compute the dot product for a range of vectors. This function is called by the threads created in dense_dot_range
Reimplemented from CDotFeatures.
Definition at line 369 of file HashedWDFeaturesTransposed.cpp.
|
virtual |
Compute the dot product for a subset of vectors. This function makes use of dense_dot alphas[i] * sparse[i]^T * w + b
sub_index | index for which to compute outputs |
num | length of index |
output | result for the given vector range |
alphas | scalars to multiply with, may be NULL |
vec | dense vector to compute dot product with |
dim | length of the dense vector |
b | bias |
Reimplemented from CDotFeatures.
Definition at line 285 of file HashedWDFeaturesTransposed.cpp.
|
virtual |
compute dot product between vector1 and vector2, appointed by their indices
vec_idx1 | index of first vector |
df | DotFeatures (of same kind) to compute dot product with |
vec_idx2 | index of second vector |
Implements CDotFeatures.
Definition at line 125 of file HashedWDFeaturesTransposed.cpp.
|
virtual |
duplicate feature object
Implements CFeatures.
Definition at line 551 of file HashedWDFeaturesTransposed.cpp.
|
virtual |
clean up iterator call this function with the iterator returned by get_first_feature
iterator | as returned by get_first_feature |
Implements CDotFeatures.
Definition at line 568 of file HashedWDFeaturesTransposed.cpp.
|
virtual |
obtain the dimensionality of the feature space
(not mix this up with the dimensionality of the input space, usually obtained via get_num_features())
Implements CDotFeatures.
Definition at line 58 of file HashedWDFeaturesTransposed.h.
|
virtual |
get feature class
Implements CFeatures.
Definition at line 146 of file HashedWDFeaturesTransposed.h.
|
virtual |
iterate over the non-zero features
call get_feature_iterator first, followed by get_next_feature and free_feature_iterator to cleanup
vector_index | the index of the vector over whose components to iterate over |
Implements CDotFeatures.
Definition at line 556 of file HashedWDFeaturesTransposed.cpp.
|
virtual |
get feature type
Implements CFeatures.
Definition at line 137 of file HashedWDFeaturesTransposed.h.
|
virtual |
Implements CSGObject.
Definition at line 221 of file HashedWDFeaturesTransposed.h.
|
virtual |
iterate over the non-zero features
call this function with the iterator returned by get_first_feature and call free_feature_iterator to cleanup
index | is returned by reference (-1 when not available) |
value | is returned by reference |
iterator | as returned by get_first_feature |
Implements CDotFeatures.
Definition at line 562 of file HashedWDFeaturesTransposed.cpp.
|
virtual |
get number of non-zero features in vector
num | which vector |
Implements CDotFeatures.
Definition at line 122 of file HashedWDFeaturesTransposed.h.
float64_t get_normalization_const | ( | ) |
get normalization constant
Definition at line 166 of file HashedWDFeaturesTransposed.h.
|
virtual |
get number of examples/vectors, possibly corresponding to the current subset
abstract base method
Implements CFeatures.
Definition at line 151 of file HashedWDFeaturesTransposed.h.
|
virtual |
get memory footprint of one feature
abstract base method
Implements CFeatures.
Definition at line 156 of file HashedWDFeaturesTransposed.h.
void set_normalization_const | ( | float64_t | n = 0 | ) |
set normalization constant
n | n=0 means automagic |
Definition at line 535 of file HashedWDFeaturesTransposed.cpp.
|
protected |
create wd kernel weighting heuristic
Definition at line 515 of file HashedWDFeaturesTransposed.cpp.
|
protected |
size of alphabet
Definition at line 247 of file HashedWDFeaturesTransposed.h.
|
protected |
degree
Definition at line 237 of file HashedWDFeaturesTransposed.h.
|
protected |
from degree
Definition at line 241 of file HashedWDFeaturesTransposed.h.
|
protected |
number of bits in hash
Definition at line 257 of file HashedWDFeaturesTransposed.h.
|
protected |
mask
Definition at line 255 of file HashedWDFeaturesTransposed.h.
|
protected |
normalization const
Definition at line 260 of file HashedWDFeaturesTransposed.h.
|
protected |
number of strings
Definition at line 245 of file HashedWDFeaturesTransposed.h.
|
protected |
partial w dim == hashsize
Definition at line 251 of file HashedWDFeaturesTransposed.h.
|
protected |
start_degree
Definition at line 239 of file HashedWDFeaturesTransposed.h.
|
protected |
length of string in vector
Definition at line 243 of file HashedWDFeaturesTransposed.h.
|
protected |
stringfeatures the wdfeatures are based on
Definition at line 231 of file HashedWDFeaturesTransposed.h.
|
protected |
pointer to transposed strings
Definition at line 234 of file HashedWDFeaturesTransposed.h.
|
protected |
w dim
Definition at line 249 of file HashedWDFeaturesTransposed.h.
|
protected |
wd weights
Definition at line 253 of file HashedWDFeaturesTransposed.h.