svdmf.sql_in File Reference

SQL functions for SVD Matrix Factorization. More...

Functions
text	svdmf_run (text input_table, text col_name, text row_name, text value, int num_features)
	Partial SVD decomposition of a sparse matrix into U and V components. More...
text	svdmf_run (text input_table, text col_name, text row_name, text value, int num_features, int num_iterations, float min_error)
	Partial SVD decomposition of a sparse matrix into U and V components. More...

Detailed Description

Date

January 2011

See also

For a brief introduction to SVD Matrix Factorization, see the module description Singular Value Decomposition.

Function Documentation

text svdmf_run	(	text	input_table,
		text	col_name,
		text	row_name,
		text	value,
		int	num_features
	)

This function takes as input the table representation of a sparse matrix and decomposes it into the specified set of most significant features of matrices of U and V matrix.

Parameters

input_table	Name of the table/view with the source data
col_name	Name of the column containing cell column number
row_name	Name of the column containing cell row number
value	Name of the column containing cell value
num_features	Rank of desired approximation

text svdmf_run	(	text	input_table,
		text	col_name,
		text	row_name,
		text	value,
		int	num_features,
		int	num_iterations,
		float	min_error
	)

This function takes as input the table representation of a sparse matrix and decomposes it into the specified set of most significant features of matrices of U and V matrix.

Parameters

input_table	Name of the table/view with the source data
col_name	Name of the column containing cell column number
row_name	Name of the column containing cell row number
value	Name of the column containing cell value
num_features	Rank of desired approximation
num_iterations	Maximum number if iterations to perform regardless of convergence
min_error	Acceptable level of error in convergence.