from data_parser import DataParser import numpy as np import matplotlib.pyplot as plt import pandas as pd parser = DataParser('europe.csv') matrix_standarized = parser.get_standarized_matrix() # Create a figure instance fig = plt.figure(1, figsize=(9, 6)) # Create an axes instance ax = fig.add_subplot(111) ## Custom x-axis labels ax.set_xticklabels([ 'Area', 'GDP', 'Inflation', 'Life.expect', 'Military', 'Pop.growth', 'Unemployment' ]) ## add patch_artist=True option to ax.boxplot() ## to get fill color bp = ax.boxplot(matrix_standarized, patch_artist=True) ## change outline color, fill color and linewidth of the boxes for box in bp['boxes']: # change outline color box.set(color='#7570b3', linewidth=2) # change fill color box.set(facecolor='#1b9e77')
from kohonen_network import Kohonen from data_parser import DataParser import seaborn as sn import matplotlib.pyplot as plt import numpy as np parser = DataParser('europe.csv') standarized_matrix = np.array(parser.get_standarized_matrix()).T k_neurons = 15 kohonen = Kohonen(standarized_matrix, k_neurons, iteration_limit=3000, eta=0.0001) kohonen.train() som_map = kohonen.construct_nodes_map() u_matrix = kohonen.build_u_matrix() countries_string = parser.parse_as_class() neurons = [[[] for x in range(k_neurons)] for y in range(k_neurons)] for i in range(len(standarized_matrix)): best_i, best_j, best_difference = kohonen.get_best_matching( standarized_matrix[i]) neurons[best_i][best_j].append(countries_string[i][0]) print("Country: {}, Best neuron: {},{}, Best difference: {}".format( countries_string[i][0], best_i, best_j, best_difference)) print('Paises por neurona') print(neurons)
from data_parser import DataParser import numpy as np import matplotlib.pyplot as plt import pandas as pd import seaborn as sn from sklearn.covariance import EmpiricalCovariance from sklearn.decomposition import PCA parser = DataParser('europe.csv') # print(parser.get_numerical_csv()) # print(np.matrix(parser.get_numerical_csv()).T) # matrix_for_correlation = np.array(parser.get_numerical_csv(), dtype=float).T matrix_for_correlation = np.array(parser.get_standarized_matrix()) print('MEDIA (Tiene que tender a 0):', np.mean(matrix_for_correlation), '| STD (Tiene que tender a 1):', np.std(matrix_for_correlation)) # print('ESTANDARIZADA',parser.get_standarized_matrix()) # 'Area','GDP','Inflation','Life.expect','Military','Pop.growth','Unemployment' matrix_for_correlation_with_keys = { 'Area': matrix_for_correlation[0], 'GDP': matrix_for_correlation[1], 'Inflation': matrix_for_correlation[2], 'Life.expect': matrix_for_correlation[3], 'Military': matrix_for_correlation[4], 'Pop.growth': matrix_for_correlation[5], 'Unemployment': matrix_for_correlation[6] } df = pd.DataFrame(matrix_for_correlation_with_keys, columns=[