get_cov_info¶

Get Covariate Information¶

Prototype¶

# at_cascade.get_cov_info
def get_cov_info(
    option_all_table      ,
    covariate_table       ,
    split_reference_table ,
) :
    assert type(option_all_table) == list
    assert type(covariate_table) == list
    assert type(split_reference_table) == list
    # ...
    assert type(cov_info) == dict
    return cov_info

option_all_table¶

This is the option_all_table as a python list of python dictionaries. This argument can’t be None.

covariate_table¶

This is the dismod_at covariate table as a python list of python dictionaries. This argument can’t be None.

split_reference_table¶

This is the split_reference_table as a python list of python dictionaries. This argument can’t be None.

cov_info¶

The return value cov_info is a dict with the following keys:

abs_covariate_id_set¶

if key is abs_covariate_id_set, cov_info[key] is a set of int. A covariate_id is in this set if and only if the corresponding covariate name is in absolute_covariates. If absolute_covariates does not appear in the option_all table, cov_info[key] is the empty set. The reference value for these absolute covariates is the same for all nodes and all values of the splitting covariate.

rel_covariate_id_set¶

if key is rel_covariate_id_set, cov_info[key] is a set of int. This is the set of covariate_id values corresponding to the relative covariates.

split_covariate_id¶

If split_reference_table is empty, this key is not present. Otherwise, for key equal to split_covariate_id, cov_info[key] is an int equal to the covariate_id corresponding to the split_covariate_name.

split_reference_list¶

If split_reference_table is empty, this key is not present. Otherwise, for key equal to split_reference_list, cov_info[key] is a list of float representation of split_reference_value in the same order as they appear in the table.

split_reference_id¶

If split_reference_table is empty, this key is not present. Otherwise, for key equal to split_reference_id, cov_info[key] is an int containing an index in the split_reference_list. The corresponding value in split_reference_list is equal to the reference value for split_covariate_name in the covariate table.