def scramble(text, path=None):
if path:
file = open(path, 'r').read()
sentences = tokenizer.tokenize(file)
sentences = tokenizer.tokenize(text)
split_sent = []
for s in sentences:
s = s.translate(str.maketrans('', '', punctuation))
split_sent.append(s.lower().split())
for sentence in split_sent:
for i in range(len(sentence)):
p = morph.parse(sentence[i])[0]
sentence[i] = p.normal_form
shuff_sent = shuffle_t(split_sent)
a = Persistent(split_sent=split_sent, min_count=5, window=5)
dgms_n = a.persistent()
b = Persistent(split_sent=shuff_sent, min_count=5, window=5)
dgms_sh = b.persistent()
return dn.wasserstein_distance(dgms_n[0], dgms_sh[0])
def scramble(text, path=None):
if path:
file = open(path, 'r').read()
sentences = tokenizer.tokenize(file)
sentences = tokenizer.tokenize(text)
split_sent = []
for s in sentences:
s = s.translate(str.maketrans('', '', punctuation))
split_sent.append(s.lower().split())
for sentence in split_sent:
for i in range(len(sentence)):
p = morph.parse(sentence[i])[0]
sentence[i] = p.normal_form
shuff_sent = shuffle_t(split_sent)
a = Persistent(split_sent=split_sent, min_count=5, window=5)
dgms_n = a.persistent()
b = Persistent(split_sent=shuff_sent, min_count=5, window=5)
dgms_sh = b.persistent()
return dn.wasserstein_distance(dgms_n[0], dgms_sh[0])
def compute_bottleneck_distance(all_seeds_rips_files,
remove_infinity=False,
compute_wass_distance=False,
use_persim=False,
M=10):
matrix = []
x = []
y = []
for file1 in all_seeds_rips_files:
print('Computing file: {}'.format(file1))
row = np.zeros(len(all_seeds_rips_files))
# example file1: LTHT/remote_data/saves/alexnet_nmp/mnist/42/pickle/8.pickle
split1_name = file1.split('/')
# print(split1_name)
seed, model_name, dataset, file1_name = split1_name[-5], split1_name[
-7], split1_name[-6], split1_name[-1]
# appending 'alexnet_nmp-mnist-42-8'
x.append(model_name + "-" + dataset + "-" + seed + "-" +
file1_name.split(".")[0])
rips1 = pickle.load(open(file1, 'rb'))
if remove_infinity:
l1 = list(rips1['dgms'][0][rips1['dgms'][0][:, 1] < np.inf])
else:
l1 = list(rips1['dgms'][0])
d1 = dion.Diagram(l1)
for i, file2 in enumerate(all_seeds_rips_files):
rips2 = pickle.load(open(file2, 'rb'))
if remove_infinity:
l2 = list(rips2['dgms'][0][rips2['dgms'][0][:, 1] < np.inf])
else:
l2 = list(rips2['dgms'][0])
d2 = dion.Diagram(l2)
if compute_wass_distance:
if use_persim:
wdist = persim.sliced_wasserstein_kernel(d1, d2, M=M)
else:
wdist = dion.wasserstein_distance(d1, d2, q=2)
row[i] = wdist
else:
if use_persim:
bdist = persim.bottleneck(d1, d2)
else:
bdist = dion.bottleneck_distance(d1, d2)
row[i] = bdist
matrix.append(row)
#
x = list(
map(
lambda y: '{}-{} seed:{}-{}'.format(
y.split('-')[0],
y.split('-')[1],
y.split('-')[2],
y.split('-')[3]), x))
return matrix, x
def wasserstein_dist_mat(pds):
diag = pds2diagrams(pds);
n = len(pds);
K = np.zeros((n, n));
for i in range(n):
print(str(i) + '/' + str(n));
for j in range(i, n):
# print str(i) + '/' + str(n) + ' (' + str(j) + ')'
K[i, j] = d.wasserstein_distance(diag[i], diag[j]);
K[j, i] = K[i, j];
return K;
def comp_row(i, p, dic, out):
n = dic['n'];
diag = dic['diag'];
z = int(n/p) + 1;
I = [x*p+i for x in range(z) if x*p+i < n];
# print(I);
start = 0;
end = 0;
for r in I:
row = np.zeros(n);
lb = r*n;
rb = (r+1)*n;
print(str(r) + '/' + str(n) + '\t(' + str(end - start) + 's)')
start = time.time();
for j in range(r, n):
# print(str(j) + '\t' + str(r) + '/' + str(n))
row[j] = d.wasserstein_distance(diag[r], diag[j]);
out[lb:rb] = row;
# print(row);
end = time.time();
def test_issue39():
dgm1 = np.loadtxt('data/issue39/dgm1.txt', delimiter=',')
dgm2 = np.loadtxt('data/issue39/dgm2.txt', delimiter=',')
dgm1 = d.Diagram(dgm1)
dgm2 = d.Diagram(dgm2)
dist = d.wasserstein_distance(dgm1,dgm2,q=5)
def wasserstein_distance(dia1, dia2, p=1):
return dionysus.wasserstein_distance(dia1, dia2, p)
def wasserstein_distance(barcode1, barcode2):
dg1, dg2 = barcode_to_diagram(barcode1), barcode_to_diagram(barcode2)
return di.wasserstein_distance(dg1, dg2)
distances = [get_distance(other_point, point) for other_point in trees]
trees = trees[np.argpartition(distances, num_points)]
samples.append(trees[:num_points]) # closest num_points
samples = np.asarray(samples)
for i in range(len(samples)):
np.savetxt('./samples/samples{:d}.csv'.format(i),
samples[i],
delimiter=',')
# Constructing the persistent homology diagrams and distance matrix
dgms_list = []
for i in range(len(samples)):
sample_set = samples[i]
simplices = diode.fill_alpha_shapes(sample_set)
f = d.Filtration(simplices)
m = d.homology_persistence(f)
dgms = d.init_diagrams(m, f)
dgms_list.append(dgms[homology_dimension])
distance_matrix = np.zeros((len(dgms_list), len(dgms_list)))
for i in range(len(dgms_list)):
for j in range(i):
# Chose Wasserstein, q=1 to emphasise many near-diagonal points
dist = d.wasserstein_distance(dgms_list[i], dgms_list[j], q=1)
distance_matrix[i, j] = dist
distance_matrix[j, i] = distance_matrix[i, j]
np.savetxt('./processed_data/dist_matrix.csv', distance_matrix, delimiter=',')
def wasserstein_distance(dia1, dia2, p=1):
return dionysus.wasserstein_distance(dia1, dia2, p)