SQL functions for elastic net regularization. More...

Functions
void	elastic_net_train (text tbl_source, text tbl_result, text col_dep_var, text col_ind_var, text regress_family, float8 alpha, float8 lambda_value, boolean standardize, text grouping_col, text optimizer, text optimizer_params, text excluded, integer max_iter, float8 tolerance)
	Interface for elastic net. More...

text	elastic_net_train ()
	Help function, to print out the supported families.

text	elastic_net_train (text family_or_optimizer)
	Help function, to print out the supported optimizer for a family or print out the parameter list for an optimizer. More...

void	elastic_net_predict (text tbl_model, text tbl_new_source, text col_id, text tbl_predict)
	Prediction and put the result in a table can be used together with General-CV. More...

float8	elastic_net_predict (text regress_family, float8[] coefficients, float8 intercept, float8[] ind_var)
	Prediction use learned coefficients for a given example. More...

float8	elastic_net_gaussian_predict (float8[] coefficients, float8 intercept, float8[] ind_var)
	Prediction for linear models use learned coefficients for a given example. More...

boolean	elastic_net_binomial_predict (float8[] coefficients, float8 intercept, float8[] ind_var)
	Prediction for logistic models use learned coefficients for a given example. More...

float8	elastic_net_binomial_prob (float8[] coefficients, float8 intercept, float8[] ind_var)
	Compute the probability of belonging to the True class for a given observation. More...

Detailed Description

See Also: For a brief introduction to elastic net, see the module description grp_lasso.

Definition in file elastic_net.sql_in.

Function Documentation

boolean elastic_net_binomial_predict	(	float8[]	coefficients,
		float8	intercept,
		float8[]	ind_var
	)

Parameters

coefficients	Logistic fitting coefficients
intercept	Logistic fitting intercept
ind_var	Features (independent variables)

returns a boolean value

Definition at line 750 of file elastic_net.sql_in.

float8 elastic_net_binomial_prob	(	float8[]	coefficients,
		float8	intercept,
		float8[]	ind_var
	)

Parameters

coefficients	Logistic fitting coefficients
intercept	Logistic fitting intercept
ind_var	Features (independent variables)

returns a double value, which is the probability of this data point being True class

Definition at line 768 of file elastic_net.sql_in.

float8 elastic_net_gaussian_predict	(	float8[]	coefficients,
		float8	intercept,
		float8[]	ind_var
	)

Parameters

coefficients	Linear fitting coefficients
intercept	Linear fitting intercept
ind_var	Features (independent variables)

returns a double value

Definition at line 732 of file elastic_net.sql_in.

Parameters

tbl_model	The result from elastic_net_train
tbl_new_source	Data table
col_id	Unique ID associated with each row
tbl_predict	Prediction result

Definition at line 670 of file elastic_net.sql_in.

Parameters

regress_family	model family
coefficients	The fitting coefficients
intercept	The fitting intercept
ind_var	Features (independent variables)

returns a double value. When regress_family is 'binomial' or 'logistic', this function returns 1 for True and 0 for False

Definition at line 692 of file elastic_net.sql_in.

Parameters

tbl_source	Name of data source table
tbl_result	Name of the table to store the results
col_ind_var	Name of independent variable column, independent variable is an array
col_dep_var	Name of dependent variable column
regress_family	Response type (gaussian or binomial)
alpha	The elastic net parameter, [0, 1]
lambda_value	The regularization parameter
standardize	Whether to normalize the variables (default True)
grouping_col	List of columns on which to apply grouping (currently only a placeholder)
optimizer	The optimization algorithm, 'fista' or 'igd'. Default is 'fista'
optimizer_params	Parameters of the above optimizer, the format is 'arg = value, ...'. Default is NULL
exclude	Which columns to exclude? Default is NULL (applicable only if col_ind_var is set as * or a column of array, column names as 'col1, col2, ...' if col_ind_var is '*'; element indices as '1,2,3, ...' if col_ind_var is a column of array)
max_iter	Maximum number of iterations to run the algorithm (default value of 10000)
tolerance	Iteration stopping criteria. Default is 1e-6

Definition at line 487 of file elastic_net.sql_in.

text elastic_net_train ( text family_or_optimizer)

Parameters

family_or_optimizer Response type, 'gaussian' or 'binomial', or optimizer type

Definition at line 652 of file elastic_net.sql_in.