runplot_figure_3.ipy

import numpy as np
import matplotlib.pyplot as plt

from pskf.tools.plot import specs as sc
from pskf.tools.plot import plotfunctions as pf
import pskf.scripts.numcomp.arrays as na
import pskf.scripts.numcomp.plot as npl
import pskf.scripts.numcomp.read as nr
import pskf.tools.run.pythonmodule as pm

# Switches
is_read = 0
is_matrix = 1
is_save = 0
is_show = 1
is_backup = 0

model_name = 'wavereal'
n_comparisons = 10000
method = 'meanc'
ensemble_sizes = [50, 70, 100, 250]

for which_methods in [[0, 1, 2, 3, 4, 5, 6]]:
    for ensemble_size in ensemble_sizes[::3]:

        # Sort which_methods
        indsorts = pf.indsort(np.sort(which_methods),
                              model_name=model_name,
                              ensemble_size=ensemble_size,
                              ensemble_sizes=ensemble_sizes,
                              )
        which_methods_sorted = [which_methods[indsorts[i]]
                                for i in range(len(which_methods))]

        for n_syn in [1, 10, 100]:

            # Read
            if is_read:
                probs_array, probs_array_name = nr.read(
                    which_methods=which_methods_sorted,
                    model=model_name,
                    ensemble_size=ensemble_size,
                    n_syn=n_syn,
                    n_comparisons=n_comparisons,
                    method=method,
                )

                np.save(probs_array_name, probs_array)
                print('Saved as ' + probs_array_name)

            # Matrix
            if is_matrix:

                # Figure
                fig = plt.figure('numcomp', figsize=[10, 10])

                # Run plot function
                ax, pic_name = npl.matrix(
                    fig.add_subplot(1, 1, 1),
                    which_methods=which_methods_sorted,
                    model=model_name,
                    ensemble_size=ensemble_size,
                    n_syn=n_syn,
                    n_comparisons=n_comparisons,
                    method=method,
                    is_text=True,
                    is_longnames=False,
                    ticklabelfontsize=20,
                    is_color_trafo=True,
                )

                # Save
                if is_save:
                    plt.savefig(pic_name)
                    print('Saved as ' + pic_name)

                # Show
                if is_show:
                    plt.show()
                else:
                    plt.clf()

# Backup
if is_backup:
    pm.py_backup(
        pm.python_scripts_dir,
        na.tag,
        "runplot_figure_3",
        "ipy",
        sc.specl(model_name,
                 '_'.join([str(i) for i in which_methods])[:],
                 '2018_08_10')
    )