----------------------------------------------
lines 6-78 of file: at_cascade/csv/pre_user.py
----------------------------------------------

{xrst_begin csv.pre_user}
{xrst_spell
  tru
}

Convert Prediction Csv Files From dismod_at Notation to User csv.fit Notation
#############################################################################

Prototype
*********
{xrst_literal
   # BEGIN_DEF
   # END_DEF
}

fit_dir
*******
Same as the csv fit :ref:`csv.fit@fit_dir` .

sim_dir
*******
Same as :ref:`csv.predict@sim_dir` .

job_table
*********
is the :ref:`create_job_table@job_table` for this cascade.

start_job_name
**************
Is the name of the job (fit) that the predictions start at.
This is a node name, followed by a period, followed by a sex.
Only this fit, and its descendants, were included in the predictions.
If this argument is None, all of the jobs (fits) were be included.

predict_job_id_list
*******************
Each element of this list is an index in the *job_table*
of a job that was included in the predictions.

node_table
**********
is the dismod_at node table for this cascade.

root_node_id
************
is the node table id for the root node of the cascade.

root_split_reference_id
***********************
is the split_reference table id for the root job of the cascade

root_database
*************
specifies the location of the dismod_at
:ref:`glossary@root_database`.

Input Prediction Files
**********************
Each job in the *predict_job_id_list* has a corresponding directory.
For prefix equal to ``fit`` , ``sam`` ,
the file *prefix*\ _prior.csv or *prefix\ _posterior.csv
must exist in each of these directories
( see :ref:`pre_one_job@Csv Output Files` ).
In addition if *sim_dir* is not None,
the file tru_prior.csv or true_posterior.csv
must also exist. These files use dismod_at notation.

Output Prediction Files
***********************
The predictions get converted to csv.predict notation; see
:ref:`csv.predict@Output Files` .

{xrst_end csv.pre_user}