def generate_groups(get_labels,
f_input='segm/segmented_curves_filtered.txt',
output=''):
for cut_value, ktotal in [(0.7, 2), (0.35, 3), (0.4, 4), (0.56, 5)]:
slopes, intervals = select_original_breakpoints(ktotal, f_input)
dois = get_dois(ktotal, f_input)
data = np.concatenate((slopes, intervals), axis=1)
data = norm(data)
labels = get_labels(data, 'single', cut_value, ktotal, output)
save_groups(dois, labels, data,
output + 'k' + str(ktotal) + '/k' + str(ktotal))
plt.savefig(filename)
if __name__ == '__main__':
sources = [
'clusters\\clusters\\clusters_ind_single_0.35_3.txt',
'clusters\\clusters\\clusters_ind_single_0.56_5.txt'
]
colors = ['tab:red', 'tab:blue', 'tab:orange', 'tab:green', 'tab:grey']
letters = ['a', 'b', 'c', 'd', 'e', 'f']
idx = 0
for N, source in zip([3, 5], sources):
labels = np.loadtxt(source, dtype=np.int)
slopes, intervals = select_original_breakpoints(
N, 'segm/segmented_curves_filtered.txt')
unique, counts = np.unique(labels, return_counts=True)
unique = unique[counts >= 10]
counts = counts[counts >= 10]
unique_idxs = np.argsort(counts)[-3:]
unique = unique[unique_idxs].tolist()
# labels = [unique.index(l) if l in unique else -1 for l in labels]
for i, label in enumerate(unique):
idxs = labels == label
slopes_i = slopes[idxs]
intervals_i = intervals[idxs]
print(label, '-> tamanho', len(slopes_i))
filename = 'ave_curve_%d_intervals_%s.pdf' % (N, letters[idx])
plot_ave_curve(slopes_i, intervals_i, filename, colors[i])
ax.set_xlabel('PCA1 %.2f' % x_exp)
ax.set_ylabel('PCA2 %.2f' % y_exp)
ax.set_zlabel('PCA3 %.2f' % z_exp)
plt.savefig(filename)
plt.clf()
if __name__ == '__main__':
sys.setrecursionlimit(100000)
f_input = 'segm/segmented_curves_html.txt'
f_output = 'data/html_by_cat/'
for N in [2, 3, 4, 5]:
print(N)
slopes, intervals = select_original_breakpoints(N, f_input)
data = np.concatenate((slopes, intervals), axis=1)
data = norm(data)
print(data.shape)
Z = linkage(data[:50000], method='single')
print('Z complete')
# plt.figure(figsize=(25, 15))
# dn = dendrogram(Z, leaf_rotation=90., leaf_font_size=8.)
# plt.ylabel('distance', fontsize=18)
# plt.xticks(fontsize=16)
# plt.yticks(fontsize=16)
# plt.savefig(f_output + 'dendrogram_%d.pdf' % N)
# plt.close()