data_include

View page source

Rows of Data Table That are Include for a Fit

Prototype

# at_cascade.data_include
def data_include(
    fit_database,
    root_database
) :
    assert type( fit_database ) == str
    assert type( root_database ) == str
    assert type( data_include_table ) == list
    if len( data_include_table ) > 0 :
        assert type( data_include_table[0] ) == dict
        assert data_include_table[0]['hold_out'] == 0
    return data_include_table

fit_database

This is the database for the job that we will be fitting. All of the dismod_at hold out commands that will be used for the fit must be executed; i.e., the hold_out column in the data_subset table is the same as will be used for the fit.

root_database

This is the root node database. It contains the dismod_at tables that are the same for all the fits in the cascade.

data_include_table

This is the rows of the data table that are included in the fit; i.e., the rows that satisfy the following conditions:

  1. The node is the node being fit or a descendant of the node being fit.

  2. All the covariates are within their maximum difference limits.

  3. The hold_out value in the data table is zero

  4. The corresponding hold_out value in the data_subset table is zero.

  5. The corresponding integrand is not in the option table hold_out list.