.. highlight:: none

[mlref] section
-------------------------------

Set options for retrieving only atomic configurations from the results of RXMC calculations.
This is used, for example, to evaluate the accuracy of neural network models and to extend the training data.
The file format is as follows.

  ::

     [mlref]
     nreplicas = 3
     ndata = 50

Input Format
^^^^^^^^^^^^^

Keywords and their values are specified by a keyword and its value in the form ``keyword = value``.
Comments can also be entered by adding # (Subsequent characters are ignored).

Key words
^^^^^^^^^^

- About replica 

  - ``nreplicas``
    
    **Format :** int (natural number)

    **Description :** The number of replicas.
  
  - ``ndata``
    
    **Format :** int (natural number)

    **Description :** The number of data (configuration) to be sampled

  - ``sampler``
    
    **Format :** string (default: "linspace")

    **Description :** The method to extract :math:`N_\text{data}` samples from :math:`N` samples generated by Monte Carlo method. 

          - "linspace"

            Extract equilispaced samples by
            ``numpy.linspace(0, N-1, num=ndata, dtype=int)``

          - "random"

            Random sampling by
            ``numpy.random.choice(range(N), size=ndata, replace=False)``

.. - Others
..
..   -  ``output_frequency``
..
..      **Format :** List of int (natural number)
..
..      **Description :**  The interval at which the placement is extracted is specified in the form of a list of [ ``nsteps`` , ``sample_frequency`` ]. Here, ``nsteps`` specifies the number of steps to be taken out of the number of configurations output by the RXMC calculation (the value of ``nsteps/sample_frequency`` in the ``[replica]`` section) and ``sample_frequency`` specifies the interval at which the configurations are taken out  (the first number of steps to be taken out is fixed at 0).  The input files for the ab initio solver corresponding to that arrangement are created in the respective folders.