def debug(folders, n_components, r = None, max_dimension = 1):
X,y = load_dataset(folders)
p = Preprocess(n_components)
X = p.fit_transform(X)
if r is None:
distances = PairwiseDistances(X.tolist())
distances = ExplicitDistances(distances)
n_samples = len(X)
r_candidates = sorted(set(np.array(distances.distances).flatten()))
for r2 in r_candidates:
print r2
cx = vietoris_rips(X.tolist(), max_dimension, r2)
cords = mds_plot(X, y)
lines_plot(cx, cords)
plt.show()
else:
cx = vietoris_rips(X.tolist(), max_dimension, r)
actual_max_dimension = len(max(cx, key=len)) - 1
for d in range(actual_max_dimension, 2, -1):
sx_d = filter_simplices(cx, d)
print "dimension", d, ":", len(sx_d), "simplices"
for i, sx in enumerate(sx_d):
print i, "..."
cords = mds_plot(X, y)
edges = list(combinations(sx, 2))
lines_plot(edges, cords, color=np.random.rand(3,))
plt.show()
def prepare_data(dataset, pca_n):
global n_classes, X, y, pp, X_tr, X_inv
n_classes = len(dataset)
X, y = load_dataset(dataset)
pp = Preprocess(pca_n)
X_tr = pp.fit_transform(X)
X_inv = pp.inverse_transform(X_tr)
ax = plt.axes([0., 0., 1., 1.])
color_set = 'bgrcmykw'
colors = [color_set[int(t)] for t in target]
plt.scatter(coords[:, 0], coords[:, 1], marker='o', s=200, c=colors)
return coords
def lines_plot(lines, cords, color='r'):
lines = [line for line in lines if len(line) == 2]
for line in lines:
if len(line) != 2: continue
[p1,p2] = line
plt.plot([cords[p1, 0], cords[p2, 0]], [cords[p1, 1], cords[p2, 1]], color=color)
if __name__ == '__main__':
from dataset import load_dataset
from preprocess import Preprocess, NoScaler
from sklearn.preprocessing import StandardScaler
X,Y = load_dataset(['../data/cup', '../data/paper', '../data/pen'])
cords = mds_plot(X, Y)
plt.savefig('./mds_no_preprocessing.png')
p = Preprocess(0.75)#, NoScaler())
X = p.fit_transform(X)
cords = mds_plot(X, Y, fig=2)
plt.savefig('./mds_preprocessing.png')
#lines = [[1,2],[2,3],[3,1]]
#lines_plot(lines, cords)
#plt.show()
from preprocess import Preprocess
def vietoris_rips(points, skeleton, max, distances = None):
"""
Generate the Vietoris-Rips complex on the given set of points in 2D.
Only simplexes up to dimension skeleton are computed.
The max parameter denotes the distance cut-off value.
The distances parameter can be used to precompute distances.
"""
if distances is None:
distances = PairwiseDistances(points)
rips = Rips(distances)
simplices = Filtration()
rips.generate(skeleton, max, simplices.append)
return [list(simplex.vertices) for simplex in simplices]
if __name__ == '__main__':
from dataset import load_dataset
x,y = load_dataset(['../data/tea_cup', '../data/spoon', '../data/apple'])
p = Preprocess(0.7)
x = p.fit_transform(x)
# vietoris_rips expect a python list.
x = x.tolist()
s = vietoris_rips(x, 4, 500)
print s
