experiments_logging.py

import matplotlib.pyplot as plt
import pandas as pd
import discretization_quality_measure as dq
import data_generation as dg
from mpl_toolkits.mplot3d import Axes3D
import data_generation as dg_old
import util
import numpy as np


def plot_disc(problem, method):
    ideal_cuts = dq.parse_cuts("ideal_disc/cut_" + problem + ".txt")
    for dist in ['ID', 'CJS']:
        for threshold in ['0.3', '0.5', '0.8']:
            # dtws_irr = []
            for irr_feat in range(11):
                fig = plt.figure()
                curr = dist + "_ORIGINAL_" + threshold + "_" + problem + ("" if irr_feat == 0 else "_" + str(irr_feat))
                cuts = dq.parse_cuts("logs4/" + curr + ".csv/cut.txt")
                if not cuts:
                    continue
                for i in range(len(ideal_cuts)):
                    ax = fig.add_subplot(len(ideal_cuts), 1, i + 1)
                    ax.hlines(1, -2.5, ideal_cuts[i][-1] + 0.5)  # Draw a horizontal line
                    ax.eventplot(ideal_cuts[i], orientation='horizontal', colors='r', linelengths = 2)
                    ax.annotate(method.__name__ + ': ' + "{0:.2f}".format(method(ideal_cuts[i], cuts[i]))
                                + ', # bins: ' + str(len(cuts[i])), xy=(-2.5, 2.2))
                    for cut in ideal_cuts[i]:
                        ax.annotate(str(cut), xy=(cut, 1), xytext=(cut-0.1, -0.2))
                    ax.hlines(1, -2.5, cuts[i][-1] + 0.5)  # Draw a horizontal line
                    ax.eventplot(cuts[i], orientation='horizontal', colors='b')
                    for j, cut in enumerate(cuts[i]):
                        ax.annotate(str(cut), xy=("{0:.1f}".format(cut), 1), xytext=(cut-0.1, 1.6  if j % 2 == 0 else 0.3))

                    plt.axis('off')
                plt.savefig(method.__name__ + "_" + curr + '.png', format='png')

def plot_data_3d(data):
    if type(data) is not pd.DataFrame:
        data = pd.DataFrame(data)
    fig = plt.figure()
    ax = fig.add_subplot(111, projection='3d')
    # data = data[np.logical_and(data[0] < 0, data[1] > 0)]

    # 3d parity problem
    color_cond = {'b': data[3] == 1,
                  'k': data[3] == 2,
                  'r': data[3] == 3,
                  'g': data[3] == 4,
                  }
    for c in color_cond:
        ax.scatter(data[0][color_cond[c]], data[1][color_cond[c]], data[2][color_cond[c]], c=c, s=1)

    ## without coloring
    # ax.scatter(data[0], data[1], data[2], c='k', s=1)

    ax.set_xlabel('X0')
    ax.set_ylabel('X1')
    ax.set_zlabel('X2')

    plt.show()

def save_plot_data_3d(f, data):
    fig = plt.figure()
    ax = fig.add_subplot(111, projection='3d')
    # data = data[np.logical_and(data[0] < 0, data[1] > 0)]

    ## 3d parity problem
    # color_cond = {'b': np.logical_and(data[0] < 0, np.logical_and(data[1] > 0, data[2] < 0)),
    #               'k': np.logical_and(data[0] < 0, np.logical_and(data[1] > 0, data[2] > 0)),
    #               'g': np.logical_and(data[0] > 0, data[1] < 0),
    #               'r': np.logical_and(data[0] < 0, data[1] < 0),
    #               'c': np.logical_and(data[0] > 0, data[1] > 0),
    #               }
    # for c in color_cond:
    #     ax.scatter(data[0][color_cond[c]], data[1][color_cond[c]], data[2][color_cond[c]], c=c, s=1)

    ## without coloring
    ax.scatter(data[0], data[1], data[2], c='k', s=1)

    ax.set_xlabel('X0')
    ax.set_ylabel('X1')
    ax.set_zlabel('X2')

    plt.savefig(f.replace('.csv', '.png'), format='png')
    plt.clf()


def plot_data_2d(data):
    plt.scatter(data[0], data[1], s=1, c='k')
    plt.xlabel("dim 0")
    plt.ylabel("dim 1")
    plt.show()


def save_plot_data_2d(f, data):
    plt.scatter(data[0], data[1], s=1, c='k')
    plt.savefig(f.replace('.csv', '.png'), format='png')
    plt.clf()


def write_out_file(problem, disc_intervals, disc_points, class_labels):
    lines = ['@relation ' + util.get_escaped_name(problem) + "\n\n"]
    counter = [1]
    for i in range(len(disc_intervals)):
        lines.append(
            '@attribute dim' + str(i) + ' {' + ','.join([str(j + counter[-1]) for j in disc_intervals[i]]) + '}\n')
        counter.append(counter[-1] + len(disc_intervals[i]))
    lines.append('@attribute class {' + ','.join(['"' + str(i) + '"' for i in class_labels.unique()]) + '}\n\n')
    lines.append('@data\n')

    for i in range(len(disc_points[0])):
        for j in range(len(disc_points)):
            lines.append(str(disc_points[j][i] + counter[j]))
            lines.append(',')
        lines.append('"' + str(class_labels[i]) + '"\n')
    return lines


def get_out_files(experiment_name, disc_intervals, disc_points, class_labels, relevant_features):
    dat_lines = []
    arff_lines = ['@relation ' + experiment_name + "\n\n"]

    counter = [1]
    for i in range(relevant_features):
        arff_lines.append(
            '@attribute dim' + str(i) + ' {' + ','.join([str(j + counter[-1]) for j in disc_intervals[i]]) + '}\n')
        counter.append(counter[-1] + len(disc_intervals[i]))
    arff_lines.append('@attribute class {' + ','.join(['"' + str(i) + '"' for i in class_labels.unique()]) + '}\n\n')
    arff_lines.append('@data\n')

    for i in range(len(disc_points[0])):
        values = [str(disc_points[j][i] + counter[j]) for j in range(relevant_features)]
        dat_line = ' '.join(values)
        dat_lines.append(dat_line + " " + str(class_labels[i]) + '\n')

        arff_line = ",".join(values)
        arff_lines.append(arff_line + ',"' + str(class_labels[i]) + '"\n')
    return dat_lines, arff_lines


def get_cut_file(disc_intervals):
    lines = []
    for i in range(len(disc_intervals)):
        lines.append('dimension ' + str(i) + ' (' + str(len(disc_intervals[i])) + ' bins)\n')
        for bin in disc_intervals[i]:
            lines.append(str(disc_intervals[i][bin][1]) + '\n')
        lines.append('-------------------------------------\n')
    return lines


def get_cuts(disc_intervals):
    return [[disc_intervals[i][bin][1] for bin in disc_intervals[i]] for i in range(len(disc_intervals))]


if __name__ == '__main__':
    # rows = 20000
    # data = np.concatenate((synthetic_cube_in_cube(rows, 2, 0), np.zeros((rows, 1))), axis=1)
    # data = pd.read_csv("synthetic_cases/blobs/3d_3_blobs_aligned.csv", delimiter=";", header=None, na_values='?')
    # data = pd.read_csv("new_cubes/cubes_10_100_03_i.csv", delimiter=";", header=None, na_values='?')
    # data = pd.read_csv("new_cubes/cubes_02_03_c.csv", delimiter=";", na_values='?', header=None)
    # data = pd.read_csv('synthetic_cases/uds_new.csv', delimiter=',', header=None)
    data = pd.DataFrame(dg.correlated_data(4000, 2, 0.1, dg.func3))
    # data = pd.DataFrame(dg_old.correlated_data(4000, 3, 0.5, dg_old.func3))
    # data = pd.DataFrame(dg.cubes(4000))
    plot_data_2d(data)

    # for f in [
    #     "2d_3_cubes_aligned_xor.csv",
    #     "2d_2_cubes_aligned.csv",
    #     "2d_2_cubes_xor.csv",
    #     # "3d_2_cubes_aligned.csv",
    #     # "3d_2_cubes_xor.csv",
    #     # "3d_3_cubes_aligned.csv",
    #     # "3d_3_cubes_aligned_xor.csv",
    #     # "3d_3_cubes_xor.csv",
    #     # "3d_4_cubes_1_aligned_xor.csv",
    #     # "3d_4_cubes_2_aligned.csv",
    #     # "3d_4_cubes_xor.csv",
    #     # "4d_2_cubes_aligned.csv",
    #     # "4d_3_cubes_aligned_xor.csv",
    #     # "4d_3_cubes_xor.csv",
    #     # "4d_4_cubes_aligned_xor.csv",
    #     # "4d_4_cubes_2_aligned.csv",
    #     # "4d_4_cubes_xor.csv",
    # ]:
    #     save_plot_data_2d(f, pd.read_csv("synthetic_cases/cubes/" + f, delimiter=";", header=None, na_values='?'))