Detailed Description

template<class ST>
class shogun::CStringFeatures< ST >

Template class StringFeatures implements a list of strings.

As this class is a template the underlying storage type is quite arbitrary and not limited to character strings, but could also be sequences of floating point numbers etc. Strings differ from matrices (cf. CSimpleFeatures) in a way that the dimensionality of the feature vectors (i.e. the strings) is not fixed; it may vary between strings.

Most string kernels require StringFeatures but a number of them actually requires strings to have same length.

When preprocessors are attached to string features they may shorten the string, but are not allowed to return strings longer than max_string_length, as some algorithms depend on this.

Also note that string features cannot currently be computed on-the-fly.

(Partly) subset access is supported for this feature type. Simple use the (inherited) set_subset(), remove_subset() functions. If done, all calls that work with features are translated to the subset. See comments to find out whether it is supported for that method

Definition at line 71 of file StringFeatures.h.

Inheritance diagram for CStringFeatures< ST >:

[legend]

List of all members.

Public Member Functions
	CStringFeatures ()
	CStringFeatures (EAlphabet alpha)
	CStringFeatures (SGStringList< ST > string_list, EAlphabet alpha)
	CStringFeatures (SGStringList< ST > string_list, CAlphabet *alpha)
	CStringFeatures (CAlphabet *alpha)
	CStringFeatures (const CStringFeatures &orig)
	CStringFeatures (CFile *loader, EAlphabet alpha=DNA)
virtual	~CStringFeatures ()
virtual void	cleanup ()
virtual void	cleanup_feature_vector (int32_t num)
virtual void	cleanup_feature_vectors (int32_t start, int32_t stop)
virtual EFeatureClass	get_feature_class ()
virtual EFeatureType	get_feature_type ()
CAlphabet *	get_alphabet ()
virtual CFeatures *	duplicate () const
SGVector< ST >	get_feature_vector (int32_t num)
void	set_feature_vector (SGVector< ST > vector, int32_t num)
void	enable_on_the_fly_preprocessing ()
void	disable_on_the_fly_preprocessing ()
ST *	get_feature_vector (int32_t num, int32_t &len, bool &dofree)
CStringFeatures< ST > *	get_transposed ()
SGString< ST > *	get_transposed (int32_t &num_feat, int32_t &num_vec)
void	free_feature_vector (ST *feat_vec, int32_t num, bool dofree)
void	free_feature_vector (SGVector< ST > feat_vec, int32_t num)
virtual ST	get_feature (int32_t vec_num, int32_t feat_num)
virtual int32_t	get_vector_length (int32_t vec_num)
virtual int32_t	get_max_vector_length ()
virtual int32_t	get_num_vectors () const
floatmax_t	get_num_symbols ()
floatmax_t	get_max_num_symbols ()
floatmax_t	get_original_num_symbols ()
int32_t	get_order ()
ST	get_masked_symbols (ST symbol, uint8_t mask)
ST	shift_offset (ST offset, int32_t amount)
ST	shift_symbol (ST symbol, int32_t amount)
virtual void	load (CFile *loader)
void	load_ascii_file (char *fname, bool remap_to_bin=true, EAlphabet ascii_alphabet=DNA, EAlphabet binary_alphabet=RAWDNA)
bool	load_fasta_file (const char *fname, bool ignore_invalid=false)
bool	load_fastq_file (const char *fname, bool ignore_invalid=false, bool bitremap_in_single_string=false)
bool	load_from_directory (char *dirname)
void	set_features (SGStringList< ST > feats)
bool	set_features (SGString< ST > *p_features, int32_t p_num_vectors, int32_t p_max_string_length)
bool	append_features (CStringFeatures< ST > *sf)
bool	append_features (SGString< ST > *p_features, int32_t p_num_vectors, int32_t p_max_string_length)
SGStringList< ST >	get_features ()
virtual SGString< ST > *	get_features (int32_t &num_str, int32_t &max_str_len)
virtual SGString< ST > *	copy_features (int32_t &num_str, int32_t &max_str_len)
virtual void	get_features (SGString< ST > *dst, int32_t num_str)
virtual void	save (CFile *writer)
virtual bool	load_compressed (char *src, bool decompress)
virtual bool	save_compressed (char *dest, E_COMPRESSION_TYPE compression, int level)
virtual int32_t	get_size ()
virtual bool	apply_preprocessor (bool force_preprocessing=false)
int32_t	obtain_by_sliding_window (int32_t window_size, int32_t step_size, int32_t skip=0)
int32_t	obtain_by_position_list (int32_t window_size, CDynamicArray< int32_t > *positions, int32_t skip=0)
bool	obtain_from_char (CStringFeatures< char > *sf, int32_t start, int32_t p_order, int32_t gap, bool rev)
template<class CT >
bool	obtain_from_char_features (CStringFeatures< CT > *sf, int32_t start, int32_t p_order, int32_t gap, bool rev)
bool	have_same_length (int32_t len=-1)
void	embed_features (int32_t p_order)
void	compute_symbol_mask_table (int64_t max_val)
void	unembed_word (ST word, uint8_t *seq, int32_t len)
ST	embed_word (ST *seq, int32_t len)
void	determine_maximum_string_length ()
virtual void	set_feature_vector (int32_t num, ST *string, int32_t len)
virtual void	get_histogram (float64_t *hist, int32_t rows, int32_t *cols, bool normalize=true)
virtual void	create_random (float64_t *hist, int32_t rows, int32_t cols, int32_t num_vec)
virtual CFeatures *	copy_subset (SGVector< index_t > indices)
virtual const char *	get_name () const
virtual void	subset_changed_post ()
template<>
EFeatureType	get_feature_type ()
template<>
EFeatureType	get_feature_type ()
template<>
EFeatureType	get_feature_type ()
template<>
EFeatureType	get_feature_type ()
template<>
EFeatureType	get_feature_type ()
template<>
EFeatureType	get_feature_type ()
template<>
EFeatureType	get_feature_type ()
template<>
EFeatureType	get_feature_type ()
template<>
EFeatureType	get_feature_type ()
template<>
EFeatureType	get_feature_type ()
template<>
EFeatureType	get_feature_type ()
template<>
EFeatureType	get_feature_type ()
template<>
bool	get_masked_symbols (bool symbol, uint8_t mask)
template<>
float32_t	get_masked_symbols (float32_t symbol, uint8_t mask)
template<>
float64_t	get_masked_symbols (float64_t symbol, uint8_t mask)
template<>
floatmax_t	get_masked_symbols (floatmax_t symbol, uint8_t mask)
template<>
bool	shift_offset (bool symbol, int32_t amount)
template<>
float32_t	shift_offset (float32_t symbol, int32_t amount)
template<>
float64_t	shift_offset (float64_t symbol, int32_t amount)
template<>
floatmax_t	shift_offset (floatmax_t symbol, int32_t amount)
template<>
bool	shift_symbol (bool symbol, int32_t amount)
template<>
float32_t	shift_symbol (float32_t symbol, int32_t amount)
template<>
float64_t	shift_symbol (float64_t symbol, int32_t amount)
template<>
floatmax_t	shift_symbol (floatmax_t symbol, int32_t amount)
template<>
bool	obtain_from_char_features (CStringFeatures< CT > *sf, int32_t start, int32_t p_order, int32_t gap, bool rev)
template<>
bool	obtain_from_char_features (CStringFeatures< CT > *sf, int32_t start, int32_t p_order, int32_t gap, bool rev)
template<>
bool	obtain_from_char_features (CStringFeatures< CT > *sf, int32_t start, int32_t p_order, int32_t gap, bool rev)
template<>
void	embed_features (int32_t p_order)
template<>
void	embed_features (int32_t p_order)
template<>
void	embed_features (int32_t p_order)
template<>
void	compute_symbol_mask_table (int64_t max_val)
template<>
void	compute_symbol_mask_table (int64_t max_val)
template<>
void	compute_symbol_mask_table (int64_t max_val)
template<>
float32_t	embed_word (float32_t *seq, int32_t len)
template<>
float64_t	embed_word (float64_t *seq, int32_t len)
template<>
floatmax_t	embed_word (floatmax_t *seq, int32_t len)
template<>
void	unembed_word (float32_t word, uint8_t *seq, int32_t len)
template<>
void	unembed_word (float64_t word, uint8_t *seq, int32_t len)
template<>
void	unembed_word (floatmax_t word, uint8_t *seq, int32_t len)
Static Public Member Functions
static ST *	get_zero_terminated_string_copy (SGString< ST > str)
Protected Member Functions
virtual ST *	compute_feature_vector (int32_t num, int32_t &len)
Protected Attributes
CAlphabet *	alphabet
int32_t	num_vectors
SGString< ST > *	features
ST *	single_string
int32_t	length_of_single_string
	length of prior single string
int32_t	max_string_length
floatmax_t	num_symbols
	number of used symbols
floatmax_t	original_num_symbols
	original number of used symbols (before higher order mapping)
int32_t	order
	order used in higher order mapping
ST *	symbol_mask_table
	order used in higher order mapping
bool	preprocess_on_get
	preprocess on-the-fly?
CCache< ST > *	feature_cache

Constructor & Destructor Documentation

CStringFeatures ( )

default constructor

Definition at line 20 of file StringFeatures.cpp.

CStringFeatures ( EAlphabet alpha )

constructor

Parameters:

alpha

alphabet (type) to use for string features

Definition at line 26 of file StringFeatures.cpp.

CStringFeatures	(	SGStringList< ST >	string_list,
		EAlphabet	alpha
	)

constructor

Parameters:

	string_list
	alpha	alphabet (type) to use for string features

Definition at line 36 of file StringFeatures.cpp.

CStringFeatures	(	SGStringList< ST >	string_list,
		CAlphabet *	alpha
	)

constructor

Parameters:

	string_list
	alpha	an actual alphabet

Definition at line 48 of file StringFeatures.cpp.

CStringFeatures ( CAlphabet * alpha )

constructor

Parameters:

alpha

alphabet to use for string features

Definition at line 60 of file StringFeatures.cpp.

CStringFeatures ( const CStringFeatures< ST > & orig )

copy constructor

Definition at line 72 of file StringFeatures.cpp.

CStringFeatures	(	CFile *	loader,
		EAlphabet	alpha = `DNA`
	)

constructor

Parameters:

	loader	File object via which to load data
	alpha	alphabet (type) to use for string features

Definition at line 111 of file StringFeatures.cpp.

~CStringFeatures ( ) [virtual]

Definition at line 126 of file StringFeatures.cpp.

Member Function Documentation

bool append_features ( CStringFeatures< ST > * sf )

append features If the given string features have a subset, only this will be copied

not possible with subset

Parameters:

sf

features to append

Returns:: if setting was successful

Definition at line 879 of file StringFeatures.cpp.

bool append_features	(	SGString< ST > *	p_features,
		int32_t	p_num_vectors,
		int32_t	p_max_string_length
	)

append features

not possible with subset

Parameters:

	p_features	features to append
	p_num_vectors	number of vectors
	p_max_string_length	maximum string length

note that p_features will be SG_FREE()'d on success

Returns:: if setting was successful

Definition at line 901 of file StringFeatures.cpp.

bool apply_preprocessor ( bool force_preprocessing = false ) [virtual]

apply preprocessor

Parameters:

force_preprocessing

if preprocssing shall be forced

Returns:: if applying was successful

Definition at line 1159 of file StringFeatures.cpp.

void cleanup ( ) [virtual]

cleanup string features.

removes any subset before

Reimplemented in CStringFileFeatures< ST >.

Definition at line 133 of file StringFeatures.cpp.

void cleanup_feature_vector ( int32_t num ) [virtual]

cleanup a single feature vector

possible with subset

Parameters:

num

number of the vector

Reimplemented in CStringFileFeatures< ST >.

Definition at line 161 of file StringFeatures.cpp.

void cleanup_feature_vectors	(	int32_t	start,
		int32_t	stop
	)			`[virtual]`

cleanup multiple feature vectors

possible with subset

Parameters:

	start	index of first vector to be cleaned
	stop	index of the last vector to be cleaned

Definition at line 176 of file StringFeatures.cpp.

ST * compute_feature_vector	(	int32_t	num,
		int32_t &	len
	)			`[protected, virtual]`

compute feature vector for sample num if target is set the vector is written to target len is returned by reference

possible with subset

Parameters:

	num	which vector
	len	length of vector

Returns:: feature vector

Definition at line 1625 of file StringFeatures.cpp.

void compute_symbol_mask_table ( int64_t max_val )

Definition at line 1863 of file StringFeatures.cpp.

void compute_symbol_mask_table ( int64_t max_val )

Definition at line 1866 of file StringFeatures.cpp.

void compute_symbol_mask_table ( int64_t max_val )

compute symbol mask table

required to access bit-based symbols

not implemented for subset

Definition at line 1356 of file StringFeatures.cpp.

void compute_symbol_mask_table ( int64_t max_val )

Definition at line 1869 of file StringFeatures.cpp.

SGString< ST > * copy_features	(	int32_t &	num_str,
		int32_t &	max_str_len
	)			`[virtual]`

copy_features

possible with subset

Parameters:

	num_str	number of strings (returned)
	max_str_len	maximal string length (returned)

Returns:: string features

Definition at line 972 of file StringFeatures.cpp.

CFeatures * copy_subset ( SGVector< index_t > indices ) [virtual]

Creates a new CFeatures instance containing copies of the elements which are specified by the provided indices.

possible with subset

Parameters:

indices

indices of feature elements to copy

Returns:: new CFeatures instance with copies of feature data

Reimplemented from CFeatures.

Definition at line 1591 of file StringFeatures.cpp.

void create_random	(	float64_t *	hist,
		int32_t	rows,
		int32_t	cols,
		int32_t	num_vec
	)			`[virtual]`

create some random strings based on normalized histogram

not possible with subset

Definition at line 1489 of file StringFeatures.cpp.

void determine_maximum_string_length ( )

determine new maximum string length

possible with subset

Definition at line 1412 of file StringFeatures.cpp.

void disable_on_the_fly_preprocessing ( )

call this to disable on the fly feature preprocessing on get_feature_vector. Useful when you manually apply preprocessors.

Definition at line 256 of file StringFeatures.cpp.

CFeatures * duplicate ( ) const [virtual]

duplicate feature object

Returns:: feature object

Implements CFeatures.

Definition at line 204 of file StringFeatures.cpp.

void embed_features ( int32_t p_order )

Definition at line 1853 of file StringFeatures.cpp.

void embed_features ( int32_t p_order )

Definition at line 1859 of file StringFeatures.cpp.

void embed_features ( int32_t p_order )

Definition at line 1856 of file StringFeatures.cpp.

void embed_features ( int32_t p_order )

embed string features in bit representation in-place

not implemented for subset

Definition at line 1302 of file StringFeatures.cpp.

floatmax_t embed_word	(	floatmax_t *	seq,
		int32_t	len
	)

Definition at line 1881 of file StringFeatures.cpp.

float32_t embed_word	(	float32_t *	seq,
		int32_t	len
	)

Definition at line 1873 of file StringFeatures.cpp.

float64_t embed_word	(	float64_t *	seq,
		int32_t	len
	)

Definition at line 1877 of file StringFeatures.cpp.

ST embed_word	(	ST *	seq,
		int32_t	len
	)

embed a single word

Parameters:

	seq	sequence of size len in a bitfield
	len

Definition at line 1399 of file StringFeatures.cpp.

void enable_on_the_fly_preprocessing ( )

call this to preprocess string features upon get_feature_vector

Definition at line 251 of file StringFeatures.cpp.

void free_feature_vector	(	SGVector< ST >	feat_vec,
		int32_t	num
	)

free feature vector

possible with subset

Parameters:

	feat_vec	feature vector to free
	num	index in feature cache, possibly from subset

Definition at line 361 of file StringFeatures.cpp.

void free_feature_vector	(	ST *	feat_vec,
		int32_t	num,
		bool	dofree
	)

free feature vector

possible with subset

Parameters:

	feat_vec	feature vector to free
	num	index in feature cache, possibly from subset
	dofree	if vector should be really deleted

Definition at line 343 of file StringFeatures.cpp.

CAlphabet * get_alphabet ( )

get alphabet used in string features

Returns:: alphabet

Definition at line 198 of file StringFeatures.cpp.

ST get_feature	(	int32_t	vec_num,
		int32_t	feat_num
	)			`[virtual]`

get feature

possible with subset

Parameters:

	vec_num	which vector
	feat_num	which feature, possibly from subset

Returns:: feature

Definition at line 379 of file StringFeatures.cpp.

EFeatureClass get_feature_class ( ) [virtual]

get feature class

Returns:: feature class STRING

Implements CFeatures.

Definition at line 194 of file StringFeatures.cpp.