Features that compute the Weighted Degreee Kernel feature space explicitly.
Definition at line 28 of file HashedWDFeatures.h.
Classes | |
struct | hashed_wd_feature_iterator |
Public Member Functions | |
CHashedWDFeatures (void) | |
CHashedWDFeatures (CStringFeatures< uint8_t > *str, int32_t start_order, int32_t order, int32_t from_order, int32_t hash_bits=12) | |
CHashedWDFeatures (const CHashedWDFeatures &orig) | |
virtual | ~CHashedWDFeatures () |
virtual int32_t | get_dim_feature_space () |
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 | 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 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 CFeatures * | duplicate () const |
virtual EFeatureType | get_feature_type () |
virtual EFeatureClass | get_feature_class () |
virtual int32_t | get_num_vectors () |
virtual int32_t | get_size () |
void | set_normalization_const (float64_t n=0) |
float64_t | get_normalization_const () |
virtual const char * | get_name () const |
Protected Member Functions | |
void | set_wd_weights () |
Protected Attributes | |
CStringFeatures< uint8_t > * | 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 |
CHashedWDFeatures | ( | void | ) |
default constructor
Definition at line 16 of file HashedWDFeatures.cpp.
CHashedWDFeatures | ( | 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 37 of file HashedWDFeatures.cpp.
CHashedWDFeatures | ( | const CHashedWDFeatures & | orig | ) |
copy constructor
Definition at line 64 of file HashedWDFeatures.cpp.
~CHashedWDFeatures | ( | ) | [virtual] |
destructor
Definition at line 80 of file HashedWDFeatures.cpp.
void add_to_dense_vec | ( | float64_t | alpha, | |
int32_t | vec_idx1, | |||
float64_t * | vec2, | |||
int32_t | vec2_len, | |||
bool | abs_val = false | |||
) | [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 164 of file HashedWDFeatures.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 118 of file HashedWDFeatures.cpp.
float64_t dot | ( | int32_t | vec_idx1, | |
CDotFeatures * | df, | |||
int32_t | vec_idx2 | |||
) | [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 86 of file HashedWDFeatures.cpp.
CFeatures * duplicate | ( | ) | const [virtual] |
duplicate feature object
Implements CFeatures.
Definition at line 249 of file HashedWDFeatures.cpp.
virtual void free_feature_iterator | ( | void * | iterator | ) | [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 157 of file HashedWDFeatures.h.
virtual int32_t get_dim_feature_space | ( | ) | [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 HashedWDFeatures.h.
virtual EFeatureClass get_feature_class | ( | ) | [virtual] |
get feature class
Implements CFeatures.
Definition at line 181 of file HashedWDFeatures.h.
virtual void* get_feature_iterator | ( | int32_t | vector_index | ) | [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 130 of file HashedWDFeatures.h.
virtual EFeatureType get_feature_type | ( | ) | [virtual] |
get feature type
Implements CFeatures.
Definition at line 172 of file HashedWDFeatures.h.
virtual const char* get_name | ( | void | ) | const [virtual] |
virtual bool get_next_feature | ( | int32_t & | index, | |
float64_t & | value, | |||
void * | iterator | |||
) | [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 146 of file HashedWDFeatures.h.
virtual int32_t get_nnz_features_for_vector | ( | int32_t | num | ) | [virtual] |
get number of non-zero features in vector
num | which vector |
Implements CDotFeatures.
Definition at line 95 of file HashedWDFeatures.h.
float64_t get_normalization_const | ( | ) |
get normalization constant
Definition at line 201 of file HashedWDFeatures.h.
virtual int32_t get_num_vectors | ( | ) | [virtual] |
get number of examples/vectors
abstract base method
Implements CFeatures.
Definition at line 186 of file HashedWDFeatures.h.
virtual int32_t get_size | ( | ) | [virtual] |
get memory footprint of one feature
abstract base method
Implements CFeatures.
Definition at line 191 of file HashedWDFeatures.h.
void set_normalization_const | ( | float64_t | n = 0 |
) |
set normalization constant
n | n=0 means automagic |
Definition at line 233 of file HashedWDFeatures.cpp.
void set_wd_weights | ( | ) | [protected] |
create wd kernel weighting heuristic
Definition at line 213 of file HashedWDFeatures.cpp.
int32_t alphabet_size [protected] |
size of alphabet
Definition at line 229 of file HashedWDFeatures.h.
int32_t degree [protected] |
degree
Definition at line 219 of file HashedWDFeatures.h.
int32_t from_degree [protected] |
from degree
Definition at line 223 of file HashedWDFeatures.h.
int32_t m_hash_bits [protected] |
number of bits in hash
Definition at line 239 of file HashedWDFeatures.h.
uint32_t mask [protected] |
mask
Definition at line 237 of file HashedWDFeatures.h.
float64_t normalization_const [protected] |
normalization const
Definition at line 242 of file HashedWDFeatures.h.
int32_t num_strings [protected] |
number of strings
Definition at line 227 of file HashedWDFeatures.h.
int32_t partial_w_dim [protected] |
partial w dim == hashsize
Definition at line 233 of file HashedWDFeatures.h.
int32_t start_degree [protected] |
start_degree
Definition at line 221 of file HashedWDFeatures.h.
int32_t string_length [protected] |
length of string in vector
Definition at line 225 of file HashedWDFeatures.h.
CStringFeatures<uint8_t>* strings [protected] |
stringfeatures the wdfeatures are based on
Definition at line 216 of file HashedWDFeatures.h.
int32_t w_dim [protected] |
w dim
Definition at line 231 of file HashedWDFeatures.h.
float64_t* wd_weights [protected] |
wd weights
Definition at line 235 of file HashedWDFeatures.h.