crf.sql_in File Reference

SQL functions for conditional random field. More...

Functions
float8[]	lincrf_lbfgs_step_transition (float8[], float8[], float8[], float8[], float8, float8, float8[])
float8[]	lincrf_lbfgs_step_merge_states (float8[] state1, float8[] state2)
float8[]	lincrf_lbfgs_step_final (float8[] state)
float8	internal_lincrf_lbfgs_converge (float8[] state)
lincrf_result	internal_lincrf_lbfgs_result (float8[] state)
aggregate float8[]	lincrf_lbfgs_step (float8[], float8[], float8[], float8, float8, float8[])
aggregate anyarray	array_union (anyarray)
integer	compute_lincrf (varchar source, varchar sparse_R, varchar dense_M, varchar sparse_M, varchar featureSize, integer tagSize, integer maxNumIterations)
integer	lincrf (varchar source, varchar sparse_R, varchar dense_M, varchar sparse_M, varchar featureSize, integer tagSize, varchar featureset, varchar crf_feature, integer maxNumIterations=20)
	Compute linear-chain crf coefficients and diagnostic statistics. More...
integer	lincrf (varchar source, varchar sparse_R, varchar dense_M, varchar sparse_M, varchar featureSize, integer tagSize, varchar featureset, varchar crf_feature)

Detailed Description

Date

July 2012

See Also

For a brief introduction to conditional random field, see the module description Conditional Random Field.

Function Documentation

aggregate anyarray array_union

(

anyarray

)

integer compute_lincrf	(	varchar	source,
		varchar	sparse_R,
		varchar	dense_M,
		varchar	sparse_M,
		varchar	featureSize,
		integer	tagSize,
		integer	maxNumIterations
	)

float8 internal_lincrf_lbfgs_converge

(

float8[]

state)

lincrf_result internal_lincrf_lbfgs_result

(

float8[]

state)

integer lincrf	(	varchar	source,
		varchar	sparse_R,
		varchar	dense_M,
		varchar	sparse_M,
		varchar	featureSize,
		integer	tagSize,
		varchar	featureset,
		varchar	crf_feature,
		integer	maxNumIterations = 20
	)

Parameters

source	Name of the source relation containing the training data
sparse_R	Name of the sparse single state feature column (of type DOUBLE PRECISION[])
dense_M	Name of the dense two state feature column (of type DOUBLE PRECISION[])
sparse_M	Name of the sparse two state feature column (of type DOUBLE PRECISION[])
featureSize	Name of feature size column (of type DOUBLE PRECISION)
tagSize	The number of tags in the tag set
featureset	The unique feature set
crf_feature	The Name of output feature table
maxNumIterations	The maximum number of iterations

Returns

a composite value:

• coef FLOAT8[] - Array of coefficients, \( \boldsymbol c \)
• log_likelihood FLOAT8 - Log-likelihood \( l(\boldsymbol c) \)
• num_iterations INTEGER - The number of iterations before the algorithm terminated
  
  A 'crf_feature' table is used to store all the features and corresponding weights

Note

This function starts an iterative algorithm. It is not an aggregate function. Source and column names have to be passed as strings (due to limitations of the SQL syntax).

integer lincrf	(	varchar	source,
		varchar	sparse_R,
		varchar	dense_M,
		varchar	sparse_M,
		varchar	featureSize,
		integer	tagSize,
		varchar	featureset,
		varchar	crf_feature
	)

aggregate float8 [] lincrf_lbfgs_step	(	float8	[],
		float8	[],
		float8	[],
		float8	,
		float8	,
		float8	[]
	)

float8 [] lincrf_lbfgs_step_final

(

float8[]

state)

float8 [] lincrf_lbfgs_step_merge_states	(	float8[]	state1,
		float8[]	state2
	)

float8 [] lincrf_lbfgs_step_transition	(	float8	[],
		float8	[],
		float8	[],
		float8	[],
		float8	,
		float8	,
		float8	[]
	)