-----------------------------------------------
lines 6-130 of file: at_cascade/csv/__init__.py
-----------------------------------------------

{xrst_begin csv.module}

The at_cascade.csv Python Module
################################

Interface Example
*****************
The python code that implements this interface to the at_cascade
:ref:`module-name` can be used as an example for other interfaces.

Notation
********

Demographer
===========
None of the data is in demographer notation.
For example,
:ref:`csv.simulate@Input Files@covariate.csv@time`
1990 means the beginning of 1990,
not the time interval from 1990 to 1991.

Rectangular Grid
================
Define a selection subset of a csv file as those rows with the same
value in column *select*.
For each value in *select* ,
a csv file is said to have a rectangular grid in age and time
if the following holds:

#. Define :math:`( a_1 , \ldots , a_M )`
   to be the vector of values in the age column for this selection subset.

#. Define :math:`( t_1 , \ldots , t_N )`
   to be the vector of values in the time column for this selection subset.

#. For :math:`m = 1 , \ldots , M`, :math:`n = 1,  \ldots , N` ,
   there is one and only one row in this selection subset with
   age equal to :math:`a_m`, and
   time equal to :math:`t_n`.

Data Type
=========
The actual data type for each entry in a csv file is a string; i.e.,
an arbitrary sequence of characters. Certain columns have further
restrictions as described below

Integer
-------
An integer value is a string represents of an integer.

Float
-----
A float value is a string that represents a floating point number.

Sex
---
A sex, or sex_name, is one of the following:
``female`` , ``both``, or ``male``.

Boolean
-------
A boolean value is either ``true`` or ``false`` .

Index Column
============
An index column for a csv file is an integer column
that has the row number corresponding to each row.
It starts with zero at the first row below the header row.
If a column name is an index column for two or more files,
rows with the same index value in the different files
correspond to each other.

Distributions
=============
Unless other wise specified, the mean and standard deviations that
simulate refers to are for a normal distribution.

sex_name2value
**************
The following dictionary maps each sex name to the corresponding sex value
{xrst_code py}'''
sex_name2value = { 'female' : -0.5, 'both' : 0.0, 'male' : 0.5 }
'''{xrst_code}

split_reference_table
*********************
The :ref:`option_all_table@split_covariate_name` is ``sex`` and
the :ref:`split_reference_table-name` has the following values for
csv.fit, :ref:`csv.simulate-name` and :ref:`csv.predict-name` :
{xrst_code py}'''
split_reference_table = [
   { 'split_reference_name' : 'female' , 'split_reference_value' : -0.5 },
   { 'split_reference_name' : 'both'   , 'split_reference_value' :  0.0 },
   { 'split_reference_name' : 'male'   , 'split_reference_value' : +0.5 },
]
'''{xrst_code}

Routines
********

.. BEGIN_SORT_THIS_LINE_PLUS_2
{xrst_toc_table
   at_cascade/csv/ancestor_fit.py
   at_cascade/csv/check_table.py
   at_cascade/csv/covariate_avg.py
   at_cascade/csv/covariate_both.py
   at_cascade/csv/covariate_same.py
   at_cascade/csv/covariate_spline.py
   at_cascade/csv/empty_str.py
   at_cascade/csv/fit.py
   at_cascade/csv/get_header.py
   at_cascade/csv/join_file.py
   at_cascade/csv/pre_one_job.py
   at_cascade/csv/pre_one_process.py
   at_cascade/csv/pre_parallel.py
   at_cascade/csv/pre_user.py
   at_cascade/csv/predict.py
   at_cascade/csv/read_table.py
   at_cascade/csv/set_truth.py
   at_cascade/csv/simulate.py
   at_cascade/csv/write_table.py
}
.. END_SORT_THIS_LINE_MINUS_2

{xrst_end csv.module}