def test_deprecations():
with pytest.warns(FutureWarning, match="deprecated in 0.22"):
gaussian_random_matrix(10, 100)
with pytest.warns(FutureWarning, match="deprecated in 0.22"):
sparse_random_matrix(10, 100)
def __init__(self, n_features=784, n_components=20, random_state=None):
super().__init__()
self.n_features = n_features
self.n_components = n_components
self.decode = torch.Tensor(
gaussian_random_matrix(self.n_components,
self.n_features,
random_state=random_state)).float()
def generate_random_projection_weights(original_dim, projected_dim):
random_projection_matrix = None
random_projection_matrix = gaussian_random_matrix(
n_components=projected_dim, n_features=original_dim).T
print("A Gaussian random weight matrix was creates with shape of {}".format(random_projection_matrix.shape))
print('Storing random projection matrix to disk...')
with open('random_projection_matrix', 'wb') as handle:
pickle.dump(random_projection_matrix, handle, protocol=pickle.HIGHEST_PROTOCOL)
return random_projection_matrix
def test_gaussian_random_matrix():
"""Check some statical properties of Gaussian random matrix"""
# Check that the random matrix follow the proper distribution.
# Let's say that each element of a_{ij} of A is taken from
# a_ij ~ N(0.0, 1 / n_components).
#
n_components = 100
n_features = 1000
A = gaussian_random_matrix(n_components, n_features, random_state=0)
assert_array_almost_equal(0.0, np.mean(A), 2)
assert_array_almost_equal(np.var(A, ddof=1), 1 / n_components, 1)
def test_gaussian_random_matrix():
# Check some statical properties of Gaussian random matrix
# Check that the random matrix follow the proper distribution.
# Let's say that each element of a_{ij} of A is taken from
# a_ij ~ N(0.0, 1 / n_components).
#
n_components = 100
n_features = 1000
A = gaussian_random_matrix(n_components, n_features, random_state=0)
assert_array_almost_equal(0.0, np.mean(A), 2)
assert_array_almost_equal(np.var(A, ddof=1), 1 / n_components, 1)
def create_estimator_no_params(self):
if self.estimator_name == 'K-means':
self.estimator = cluster.KMeans(n_jobs=self.n_jobs)
elif self.estimator_name == 'EM':
self.estimator = mixture.GaussianMixture()
elif self.estimator_name == 'PCA':
self.estimator = decomposition.PCA()
elif self.estimator_name == 'ICA':
self.estimator = decomposition.FastICA()
elif self.estimator_name == 'Random_Projection':
self.estimator = random_projection.gaussian_random_matrix()
elif self.estimator_name == 'Dictionary_Learning':
self.estimator = decomposition.DictionaryLearning()
def __init__(self, patch_size, nc, n_components, stride=1):
super(RandomProjection, self).__init__()
randmat = random_projection.gaussian_random_matrix(
n_components, nc * patch_size * patch_size)
self.main = nn.Conv2d(nc,
n_components,
patch_size,
stride=stride,
padding=0,
bias=False)
randmat = randmat.reshape(*self.main.weight.shape).astype(np.float32)
self.main.weight = nn.Parameter(torch.from_numpy(randmat))
for p in self.parameters():
p.requires_grad = False
def generate_random_projection_weights(original_dim, projected_dim, output_dir):
"""Generates a Gaussian random projection weights matrix."""
random_projection_matrix = None
if projected_dim and original_dim > projected_dim:
random_projection_matrix = gaussian_random_matrix(
n_components=projected_dim, n_features=original_dim).T
print('A Gaussian random weight matrix was creates with shape of {}'.format(
random_projection_matrix.shape))
print('Storing random projection matrix to disk...')
output_file_path = os.path.join(output_dir, _RANDOM_PROJECTION_FILENAME)
with open(output_file_path, 'wb') as handle:
pickle.dump(
random_projection_matrix, handle, protocol=pickle.HIGHEST_PROTOCOL)
print('Random projection matrix saved to disk.')
return random_projection_matrix
