131 lines
4.6 KiB
Python
131 lines
4.6 KiB
Python
import numpy as np
|
|
from scipy import linalg
|
|
|
|
|
|
def get_metric_statistics(values, replication_times):
|
|
mean = np.mean(values, axis=0)
|
|
std = np.std(values, axis=0)
|
|
conf_interval = 1.96 * std / np.sqrt(replication_times)
|
|
return mean, conf_interval
|
|
|
|
|
|
# (X - X_train)*(X - X_train) = -2X*X_train + X*X + X_train*X_train
|
|
def euclidean_distance_matrix(matrix1, matrix2):
|
|
"""
|
|
Params:
|
|
-- matrix1: N1 x D
|
|
-- matrix2: N2 x D
|
|
Returns:
|
|
-- dist: N1 x N2
|
|
dist[i, j] == distance(matrix1[i], matrix2[j])
|
|
"""
|
|
assert matrix1.shape[1] == matrix2.shape[1]
|
|
d1 = -2 * np.dot(matrix1, matrix2.T) # shape (num_test, num_train)
|
|
d2 = np.sum(np.square(matrix1), axis=1, keepdims=True) # shape (num_test, 1)
|
|
d3 = np.sum(np.square(matrix2), axis=1) # shape (num_train, )
|
|
dists = np.sqrt(d1 + d2 + d3) # broadcasting
|
|
return dists
|
|
|
|
|
|
def calculate_top_k(mat, top_k):
|
|
size = mat.shape[0]
|
|
gt_mat = np.expand_dims(np.arange(size), 1).repeat(size, 1)
|
|
bool_mat = (mat == gt_mat)
|
|
correct_vec = False
|
|
top_k_list = []
|
|
for i in range(top_k):
|
|
# print(correct_vec, bool_mat[:, i])
|
|
correct_vec = (correct_vec | bool_mat[:, i])
|
|
# print(correct_vec)
|
|
top_k_list.append(correct_vec[:, None])
|
|
top_k_mat = np.concatenate(top_k_list, axis=1)
|
|
return top_k_mat
|
|
|
|
|
|
def calculate_activation_statistics(activations):
|
|
"""
|
|
Params:
|
|
-- activation: num_samples x dim_feat
|
|
Returns:
|
|
-- mu: dim_feat
|
|
-- sigma: dim_feat x dim_feat
|
|
"""
|
|
mu = np.mean(activations, axis=0)
|
|
cov = np.cov(activations, rowvar=False)
|
|
return mu, cov
|
|
|
|
def calculate_frechet_distance(mu1, sigma1, mu2, sigma2, eps=1e-6):
|
|
"""Numpy implementation of the Frechet Distance.
|
|
The Frechet distance between two multivariate Gaussians X_1 ~ N(mu_1, C_1)
|
|
and X_2 ~ N(mu_2, C_2) is
|
|
d^2 = ||mu_1 - mu_2||^2 + Tr(C_1 + C_2 - 2*sqrt(C_1*C_2)).
|
|
Stable version by Dougal J. Sutherland.
|
|
Params:
|
|
-- mu1 : Numpy array containing the activations of a layer of the
|
|
inception net (like returned by the function 'get_predictions')
|
|
for generated samples.
|
|
-- mu2 : The sample mean over activations, precalculated on an
|
|
representative data set.
|
|
-- sigma1: The covariance matrix over activations for generated samples.
|
|
-- sigma2: The covariance matrix over activations, precalculated on an
|
|
representative data set.
|
|
Returns:
|
|
-- : The Frechet Distance.
|
|
"""
|
|
|
|
mu1 = np.atleast_1d(mu1)
|
|
mu2 = np.atleast_1d(mu2)
|
|
|
|
sigma1 = np.atleast_2d(sigma1)
|
|
sigma2 = np.atleast_2d(sigma2)
|
|
|
|
assert mu1.shape == mu2.shape, \
|
|
'Training and test mean vectors have different lengths'
|
|
assert sigma1.shape == sigma2.shape, \
|
|
'Training and test covariances have different dimensions'
|
|
|
|
diff = mu1 - mu2
|
|
|
|
# Product might be almost singular
|
|
covmean, _ = linalg.sqrtm(sigma1.dot(sigma2), disp=False)
|
|
if not np.isfinite(covmean).all():
|
|
msg = ('fid calculation produces singular product; '
|
|
'adding %s to diagonal of cov estimates') % eps
|
|
print(msg)
|
|
offset = np.eye(sigma1.shape[0]) * eps
|
|
covmean = linalg.sqrtm((sigma1 + offset).dot(sigma2 + offset))
|
|
|
|
# Numerical error might give slight imaginary component
|
|
if np.iscomplexobj(covmean):
|
|
if not np.allclose(np.diagonal(covmean).imag, 0, atol=1e-3):
|
|
m = np.max(np.abs(covmean.imag))
|
|
raise ValueError('Imaginary component {}'.format(m))
|
|
covmean = covmean.real
|
|
|
|
tr_covmean = np.trace(covmean)
|
|
|
|
return (diff.dot(diff) + np.trace(sigma1) +
|
|
np.trace(sigma2) - 2 * tr_covmean)
|
|
|
|
|
|
def calculate_diversity(activation, diversity_times):
|
|
assert len(activation.shape) == 2
|
|
assert activation.shape[0] > diversity_times
|
|
num_samples = activation.shape[0]
|
|
|
|
first_indices = np.random.choice(num_samples, diversity_times, replace=False)
|
|
second_indices = np.random.choice(num_samples, diversity_times, replace=False)
|
|
dist = linalg.norm(activation[first_indices] - activation[second_indices], axis=1)
|
|
return dist.mean()
|
|
|
|
|
|
def calculate_multimodality(activation, multimodality_times):
|
|
assert len(activation.shape) == 3
|
|
assert activation.shape[1] > multimodality_times
|
|
num_per_sent = activation.shape[1]
|
|
|
|
first_dices = np.random.choice(num_per_sent, multimodality_times, replace=False)
|
|
second_dices = np.random.choice(num_per_sent, multimodality_times, replace=False)
|
|
dist = linalg.norm(activation[:, first_dices] - activation[:, second_dices], axis=2)
|
|
return dist.mean()
|