def test_log_responsibilities(self):
"""
Test the log responsibilities with the help of Sklearn.
"""
N = 16384
S = 2048
D = 128
means = torch.randn(S, D)
covs = torch.rand(S)
x = torch.randn(N, D)
prior = torch.rand(S)
prior /= prior.sum()
mixture = GaussianMixture(S, covariance_type='spherical')
mixture.means_ = means.numpy()
mixture.precisions_cholesky_ = np.sqrt(1 / covs.numpy())
mixture.weights_ = prior.numpy()
# pylint: disable=protected-access
_, expected = mixture._estimate_log_prob_resp(x.numpy())
expected = torch.from_numpy(expected)
probs = log_normal(x, means, covs, 'spherical')
predicted = log_responsibilities(probs, prior)
self.assertTrue(
torch.allclose(expected, predicted, atol=1e-03, rtol=1e-05))
def log_prob(self, x, t, feature, samples):
observations = x[:, :t]
p_s_past = {} #p(s_t-1|X_{0:t-1})
for st in self.states:
p_s_past[st], _, _ = fwd_bkw(observations, self.states,
self.start_probability,
self.transition_probability,
self.emission_probability, st)
p_currstate_past = {}
for s in self.states:
p_currstate_past[s] = 0.
for curr_state in self.states:
for st in self.states:
p_currstate_past[curr_state] += self.transition_probability[
curr_state][st] * p_s_past[st]
gmm = GaussianMixture(n_components=len(self.states),
covariance_type='full')
gmm.fit(np.random.randn(10, observations.shape[0]))
gmm.weights_ = list(p_currstate_past.values())
gmm.means_ = np.array(self.mean)
gmm.covariances_ = np.array(self.cov)
for i in range(2):
gmm.precisions_[i] = np.linalg.inv(gmm.covariances_[i])
gmm.precisions_cholesky_[i] = np.linalg.cholesky(
gmm.covariances_[i])
return gmm.score_samples(samples)
def sample_gaussian_mixture(pis, sigmas, mus, num_samples, n_features):
"""
return: array of size (batch_size,num_samples*n_features) containing samples
taken from the gaussian mixture parameratized by pis, sigmas, mus
e.g
input 1 [[ s1_f1,s1_f2,s1_f3 | s2_f1, s2_f2, s2_f3 |.... ],
input 2 [ s1_f1,s1_f2,s1_f3 | s2_f1, s2_f2, s2_f3 |.... ],
input 3 [ s1_f1,s1_f2,s1_f3 | s2_f1, s2_f2, s2_f3 |.... ],
. [...............................................],
input n [ s1_f1,s1_f2,s1_f3 | s2_f1, s2_f2, s2_f3 |.... ]]
"""
# Gaussian PDF parameters
batch_size = pis.shape[0]
num_mixtures = pis.shape[1]
samples = np.zeros((batch_size, num_samples * n_features))
gmm = GaussianMixture(n_components=num_mixtures,
covariance_type='spherical')
gmm.fit(np.random.rand(10, 1)) # Now it thinks it is trained
for i in range(batch_size):
gmm.weights_ = pis[i]
gmm.means_ = mus[i].reshape(num_mixtures, n_features)
gmm.covariances_ = np.expand_dims(sigmas[i], axis=1)**2
sample = gmm.sample(num_samples)
samples[i] = np.ravel(sample[0])
return Variable(torch.from_numpy(samples))
def get_P_of_Data_Given_Param(means,
covs,
weights,
X,
method='scipy'): # P(Data | Param)
samples = zip(range(0, len(X)), X)
p = {}
if method == 'scipy':
g = [
multivariate_normal(mean=means[k],
cov=covs[k],
allow_singular=False)
for k in range(0, len(weights))
]
gaussians = {}
for index, x in samples:
gaussians[index] = np.array([g_k.pdf(x) for g_k in g])
for index, x in samples:
probabilities = np.multiply(gaussians[index], weights)
probabilities = probabilities / np.sum(probabilities)
p[index] = probabilities
else:
gmm = GaussianMixture(n_components=len(weights),
covariance_type='diag').fit(X)
gmm.precisions_cholesky_ = 1 # crude way to make GMM think that model is fit
gmm.means_ = means
gmm.covariances_ = covs
gmm.weights_ = weights
for index, x in samples:
x = x.reshape(1, -1)
likelihood_ratio = gmm.predict_proba(x.reshape(1, -1))
# likelihood_ratio = likelihood_ratio / np.sum(likelihood_ratio)
p[index] = likelihood_ratio
return p
def toGaussianMixture(self):
g = GaussianMixture(self.n)
g.fit(np.random.rand(2 * self.n).reshape((-1, 1)))
g.weights_ = np.array(self.weights)
g.means_ = np.array(self.means)[:, np.newaxis]
g.covariances_ = np.array(self.covariances)[:, np.newaxis, np.newaxis]
return g
def cause(n, k, p1, p2):
g = GaussianMixture(k)
g.means_ = p1 * np.random.randn(k, 1)
g.covars_ = np.power(abs(p2 * np.random.randn(k, 1) + 1), 2)
g.weights_ = abs(np.random.rand(k, 1))
g.weights_ = g.weights_ / sum(g.weights_)
return scale(g.sample(n))
def get_3d_grid_gmm(subdivisions=[5, 5, 5], variance=0.04):
"""
Compute the weight, mean and covariance of a gmm placed on a 3D grid
:param subdivisions: 2 element list of number of subdivisions of the 3D space in each axes to form the grid
:param variance: scalar for spherical gmm.p
:return gmm: gmm: instance of sklearn GaussianMixture (GMM) object Gauassian mixture model
"""
# n_gaussians = reduce(lambda x, y: x*y,subdivisions)
n_gaussians = np.prod(np.array(subdivisions))
step = [
1.0 / (subdivisions[0]), 1.0 / (subdivisions[1]),
1.0 / (subdivisions[2])
]
means = np.mgrid[step[0] - 1:1.0 - step[0]:complex(0, subdivisions[0]),
step[1] - 1:1.0 - step[1]:complex(0, subdivisions[1]),
step[2] - 1:1.0 - step[2]:complex(0, subdivisions[2])]
means = np.reshape(means, [3, -1]).T
covariances = variance * np.ones_like(means)
weights = (1.0 / n_gaussians) * np.ones(n_gaussians)
gmm = GaussianMixture(n_components=n_gaussians, covariance_type='diag')
gmm.weights_ = weights
gmm.covariances_ = covariances
gmm.means_ = means
from sklearn.mixture.gaussian_mixture import _compute_precision_cholesky
gmm.precisions_cholesky_ = _compute_precision_cholesky(covariances, 'diag')
return gmm
def test_fit(self):
expected_means = np.array([-55., 0., 7.])
expected_stds = np.array([3., .5, 1.])
from sklearn.mixture import GaussianMixture
gmm = GaussianMixture(n_components=3, means_init=expected_means)
gmm.means_ = expected_means[..., None]
gmm.covariances_ = np.array(expected_stds[..., None, None])
gmm.weights_ = np.array([1 / 3, 1 / 3, 1 / 3])
obs = gmm.sample(100000 + np.random.randint(-3, 3))[0].squeeze()
init = deeptime.markov.hmm.init.gaussian.from_data(obs,
n_hidden_states=3,
reversible=True)
hmm_est = deeptime.markov.hmm.MaximumLikelihoodHMM(init)
hmm = hmm_est.fit(obs).fetch_model()
np.testing.assert_array_almost_equal(
hmm.transition_model.transition_matrix, np.eye(3), decimal=3)
m = hmm.output_model
for mean, sigma in zip(m.means, m.sigmas):
# find the mean closest to this one (order might have changed)
mean_ix = np.argmin(np.abs(expected_means - mean))
np.testing.assert_almost_equal(mean,
expected_means[mean_ix],
decimal=1)
np.testing.assert_almost_equal(sigma * sigma,
expected_stds[mean_ix],
decimal=1)
def cluster(data, num_of_clusters):
x = data[[0, 1]]
y = data['label']
means = []
for clus in range(num_of_clusters):
clus_x = data[(data['label'] == clus)]
#print 'clus: %d' % clus
#print clus_x
means.append([clus_x[0].mean(), clus_x[1].mean()])
# print 'x'
# print x
# print 'y'
# print y
# clusterer = KMeans(n_clusters=num_of_clusters)
# clusterer = GaussianMixture(n_components=num_of_clusters, means_init=means)
clusterer = GaussianMixture(n_components=num_of_clusters)
clusterer.fit(x, y)
print 'clusterer means: '
print clusterer.means_
for clus in range(num_of_clusters):
xx, yy = clusterer.means_[clus]
plt.scatter([xx], [yy], c=colors[clus], marker="s", s=200, alpha=0.5)
means = np.array(means)
print 'computer means:'
print means
#print pd.DataFrame(means)
clusterer.means_ = means
print 'clusterer new means: '
print clusterer.means_
for clus in range(num_of_clusters):
xx, yy = clusterer.means_[clus]
plt.scatter([xx], [yy], c=colors[clus], marker="d", s=200, alpha=0.5)
preds = clusterer.predict(x)
data['pred'] = preds
draw_features_cluster(data, num_of_clusters)
def create_random_gmm(n_mix, n_features, covariance_type, prng=0):
prng = check_random_state(prng)
g = GaussianMixture(n_mix, covariance_type=covariance_type)
g.means_ = prng.randint(-20, 20, (n_mix, n_features))
g.covars_ = make_covar_matrix(covariance_type, n_mix, n_features)
g.weights_ = normalized(prng.rand(n_mix))
return g
def gmm_scale(gmm, shift=None, scale=None, reverse=False, params=None):
"""
Apply scaling factors to GMM instances.
Parameters
----------
gmm : GaussianMixture
GMM instance to be scaled.
shift : int, float, optional
Shift for the entire model. Default is 0 (no shift).
scale : int, float, optional
Scale for all components. Default is 1 (no scale).
reverse : bool, optional
Whether the GMM should be reversed.
params
GaussianMixture params for initialization of new instance.
Returns
-------
GaussianMixture
Modified GMM instance.
"""
# Fetch parameters if not supplied
if params is None:
# noinspection PyUnresolvedReferences
params = gmm.get_params()
# Instantiate new GMM
gmm_new = GaussianMixture(**params)
# Create scaled fitted GMM model
gmm_new.weights_ = gmm.weights_
# Apply shift if set
gmm_new.means_ = gmm.means_ + shift if shift is not None else gmm.means_
# Apply scale
if scale is not None:
gmm_new.means_ /= scale
gmm_new.covariances_ = gmm.covariances_ / scale ** 2 if scale is not None else gmm.covariances_
gmm_new.precisions_ = np.linalg.inv(gmm_new.covariances_) if scale is not None else gmm.precisions_
gmm_new.precisions_cholesky_ = np.linalg.cholesky(gmm_new.precisions_) if scale is not None \
else gmm.precisions_cholesky_
# Reverse if set
if reverse:
gmm_new.means_ *= -1
# Add converged attribute if available
if gmm.converged_:
gmm_new.converged_ = gmm.converged_
# Return scaled GMM
return gmm_new
def computeProb(mfcc):
# sil
sil_mean = np.loadtxt('sil_mean.txt')
sil_variance = np.loadtxt('sil_variance.txt')
sil_weight = np.loadtxt('sil_weight.txt')
sil_gmm = GaussianMixture(128, covariance_type="diag")
sil_precisions_cholesky_ = gaussian_mixture._compute_precision_cholesky(
sil_variance, "diag")
sil_gmm.means_ = sil_mean
sil_gmm.weights_ = sil_weight
sil_gmm.precisions_cholesky_ = sil_precisions_cholesky_
sil_result = np.dot(sil_gmm.predict_proba(mfcc), sil_weight.reshape(-1, 1))
# speech
speech_mean = np.loadtxt('speech_mean.txt')
speech_variance = np.loadtxt('speech_variance.txt')
speech_weight = np.loadtxt('speech_weight.txt')
speech_gmm = GaussianMixture(128, covariance_type="diag")
speech_precisions_cholesky_ = gaussian_mixture._compute_precision_cholesky(
speech_variance, "diag")
speech_gmm.means_ = speech_mean
speech_gmm.weights_ = speech_weight
speech_gmm.precisions_cholesky_ = speech_precisions_cholesky_
speech_result = np.dot(speech_gmm.predict_proba(mfcc),
speech_weight.reshape(-1, 1))
# noise
noise_mean = np.loadtxt('noise_mean.txt')
noise_variance = np.loadtxt('noise_variance.txt')
noise_weight = np.loadtxt('noise_weight.txt')
noise_gmm = GaussianMixture(128, covariance_type="diag")
noise_precisions_cholesky_ = gaussian_mixture._compute_precision_cholesky(
noise_variance, "diag")
noise_gmm.means_ = noise_mean
noise_gmm.weights_ = noise_weight
noise_gmm.precisions_cholesky_ = noise_precisions_cholesky_
noise_result = np.dot(noise_gmm.predict_proba(mfcc),
noise_weight.reshape(-1, 1))
return sil_result, speech_result, noise_result
def gmm_cause(points, k=2, p1=3, p2=4):
"""Init a root cause with a Gaussian Mixture Model w/ a spherical covariance type."""
g = GMM(k, covariance_type="spherical")
g.fit(np.random.randn(300, 1))
g.means_ = p1 * np.random.randn(k, 1)
g.covars_ = np.power(abs(p2 * np.random.randn(k, 1) + 1), 2)
g.weights_ = abs(np.random.rand(k))
g.weights_ = g.weights_ / sum(g.weights_)
return g.sample(points)[0].reshape(-1)
def return_copy(self):
'''If any trouble be sure that assignation of
means and weights is done copying through assignation
'''
copy_tmp = GMM(n_components=self.n_components)
copy_tmp.covariances_ = self.covariances_ #_get_covars()
copy_tmp.means_ = self.means_
copy_tmp.weights_ = self.weights_
return copy_tmp
def multimod_emd_from_gmm(means, sigmas, weights):
means_stacked = np.concatenate(means, axis=0)[:, :, 0, 0]
sigmas_stacked = np.concatenate(sigmas, axis=0)[:, :, 0, 0]
weights_stacked = np.concatenate(weights, axis=0)[:, 0, 0, 0]
gmm = GaussianMixture(n_components=4, covariance_type='diag')
gmm_vars = 2 * sigmas_stacked * sigmas_stacked
precisions_cholesky = _compute_precision_cholesky(gmm_vars, 'diag')
gmm.weights_ = weights_stacked
gmm.means_ = means_stacked
gmm.precisions_cholesky_ = precisions_cholesky
gmm.covariances_ = gmm_vars
y_sampled, _ = gmm.sample(1000)
return wemd_from_pred_samples(y_sampled)
def get_multimodality_score(means, sigmas, weights):
means_stacked = np.concatenate(means, axis=0)[:, :, 0, 0]
sigmas_stacked = np.concatenate(sigmas, axis=0)[:, :, 0, 0]
weights_stacked = np.concatenate(weights, axis=0)[:, 0, 0, 0]
gmm = GaussianMixture(n_components=4, covariance_type='diag')
vars = 2 * sigmas_stacked * sigmas_stacked
precisions_cholesky = _compute_precision_cholesky(vars, 'diag')
gmm.weights_ = weights_stacked
gmm.means_ = means_stacked
gmm.precisions_cholesky_ = precisions_cholesky
gmm.covariances_ = vars
gmm_uni = GaussianMixture(n_components=1, covariance_type='diag')
argmax = np.argmax(gmm.weights_)
gmm_uni.means_ = gmm.means_[argmax, :].reshape([1, 2])
gmm_uni.covariances_ = gmm.covariances_[argmax, :].reshape([1, 2])
gmm_uni.precisions_cholesky_ = gmm.precisions_cholesky_[argmax, :].reshape(
[1, 2])
gmm_uni.weights_ = np.array([1]).reshape([1])
Z_uni = compute_histogram_gmm(gmm_uni)
Z = compute_histogram_gmm(gmm)
ratio = computeWEMD(Z, Z_uni)
return ratio
def create_sklearn_gmm(weights, mean_tensor, cov_tensor, random_state=0):
n_components = len(weights)
gmm = GaussianMixture(n_components=n_components,
covariance_type='full',
random_state=random_state)
gmm.weights_ = weights.numpy()
gmm.means_ = mean_tensor.numpy()
gmm.covariances_ = cov_tensor.numpy()
gmm.precisions_ = np.array(
[np.linalg.inv(cov) for cov in gmm.covariances_])
gmm.precisions_cholesky_ = np.array(
[np.linalg.cholesky(prec) for prec in gmm.precisions_])
return gmm
def jsd_diss(self, w1, mu1, cov1, w2, mu2, cov2):
"""
Calculates Jensen-Shannon divergence of two gmm's
:param gmm_p: mixture.GaussianMixture
:param gmm_q: mixture.GaussianMixture
:param sample_count: number of monte carlo samples to use
:return: Jensen-Shannon divergence
"""
gmm_p = GaussianMixture(n_components=n_components,
covariance_type="full")
gmm_p.weights_ = w1
gmm_p.covariances_ = cov1
gmm_p.means_ = mu1
gmm_p.n_components = 1
gmm_p.precisions_cholesky_ = _compute_precision_cholesky(cov1, "full")
gmm_q = GaussianMixture(n_components=n_components,
covariance_type="full")
gmm_q.weights_ = w2
gmm_q.covariances_ = cov2
gmm_q.means_ = mu2
gmm_q.n_components = 1
gmm_q.precisions_cholesky_ = _compute_precision_cholesky(cov2, "full")
X = gmm_p.sample(sample_count)[0]
log_p_X = gmm_p.score_samples(X)
log_q_X = gmm_q.score_samples(X)
log_mix_X = np.logaddexp(log_p_X, log_q_X)
Y = gmm_q.sample(sample_count)[0]
log_p_Y = gmm_p.score_samples(Y)
log_q_Y = gmm_q.score_samples(Y)
log_mix_Y = np.logaddexp(log_p_Y, log_q_Y)
# black magic?
return (log_p_X.mean() -
(log_mix_X.mean() - np.log(2)) + log_q_Y.mean() -
(log_mix_Y.mean() - np.log(2))) / 2
def generate_equal_weight_GMM(H_mu, H_var, covariance_type='diag'):
n_components = len(H_mu)
weights_init = len(H_mu) * [1. / len(H_mu)]
GMM = GaussianMixture(n_components=n_components,
covariance_type=covariance_type, n_init=0, weights_init=None, means_init=None,
precisions_init=None, random_state=None, warm_start=True, verbose=0,
verbose_interval=10)
GMM.weights_ = weights_init
GMM.means_ = H_mu
GMM.covariances_ = H_var
GMM.precisions_cholesky_ = _compute_precision_cholesky(
H_var, covariance_type)
return GMM
def gmm_loglik(y,pi,mu,sigma,K):
model = GaussianMixture(K, covariance_type = 'diag')
model.fit(y)
N = np.shape(mu)[0]
N_test = np.shape(y)[0]
ll_test = np.zeros(N)
for i in (range(N)):
model.means_ = mu[i,:]
model.covariances_ = sigma[i,:]**2
model.precisions_ = 1/(sigma[i,:]**2)
model.weights_ = pi[i,:]
model.precisions_cholesky_ = _compute_precision_cholesky(model.covariances_, model.covariance_type)
ll_test[i] = model.score(y)
return ll_test*N_test
def generate_params_gmm(weight, mean, cov, use_cdf=False):
gmm = GaussianMixture(n_components=weight.size)
gmm.weights_ = weight
gmm.means_ = mean
gmm.covariances_ = cov
# Pass the fit check
gmm.precisions_cholesky_ = None
params = gmm.sample()[0][0]
if use_cdf:
params = ndtr(params)
else:
params = np.clip(params, 0, 1)
params = params.tolist()
return params
def visualize_latent_space(file_name,
z_mean,
z_std,
x_label='$\\mathbf{z}$',
y_label='pdf',
show=False):
"""Visualizes approximation ability.
Args:
file_name: File name without extension.
z_mean: ndarray (N, dZ) with latent states.
z_std: ndarray (N, dZ, dZ) with std of latent states.
x_label: ALbel of the x-axis.
y_label: ALbel of the y-axis.
show: Display generated plot. This is a blocking operation.
"""
fig = plt.figure()
ax1 = fig.add_subplot(111)
ax1.set_xlabel(x_label)
ax1.set_ylabel(y_label)
ax1.grid(linestyle=':')
N, dZ = z_mean.shape
xs = np.linspace(-3, 3, 1000).reshape(1000, 1)
plt.plot(xs,
sp.stats.norm.pdf(xs),
color="black",
linestyle=":",
label='$\\mathcal{N}(0,1)$')
for dim in range(dZ):
# Fit GMM by hand
gmm = GaussianMixture(N)
gmm.means_ = z_mean[:, dim].reshape(N, 1)
gmm.precisions_cholesky_ = (1 / z_std[:, dim]).reshape(N, 1, 1)
gmm.weights_ = np.ones(N) / N
ax1.plot(xs,
np.exp(gmm.score_samples(xs)),
linewidth=1,
label='$\\mathbf{z}[%d]$' % dim)
ax1.legend()
if file_name is not None:
fig.savefig(file_name + ".pdf", bbox_inches='tight', pad_inches=0)
if show:
plt.show()
plt.close(fig)
def _sample_rows_same(self, X):
""" uses efficient sklearn implementation to sample from gaussian mixture -> only works if all rows of X are the same"""
weights, locs, scales = self._get_mixture_components(np.expand_dims(X[0], axis=0))
# make sure that sum of weights < 1
weights = weights.astype(np.float64)
weights = weights / np.sum(weights)
gmm = GaussianMixture(n_components=self.n_centers, covariance_type='diag', max_iter=5, tol=1e-1)
gmm.fit(np.random.normal(size=(100,self.ndim_y))) # just pretending a fit
# overriding the GMM parameters with own params
gmm.converged_ = True
gmm.weights_ = weights[0]
gmm.means_ = locs[0]
gmm.covariances_ = scales[0]
y_sample, _ = gmm.sample(X.shape[0])
assert y_sample.shape == (X.shape[0], self.ndim_y)
return X, y_sample
def fit(self, df):
if len(df.columns) > 2:
return print('error: data should have up to 2-dimension')
self.data = df
if self.use_kmeans_init:
gm = GaussianMixture(n_components=self.no_clusters, random_state=0)
gm.means_ = init_k_means(df, no_clusters=self.no_clusters)
print(f'k-means clustering initialize: {gm.means_}')
gm.fit(df)
else:
gm = GaussianMixture(n_components=self.no_clusters,
random_state=0).fit(df)
print(f'clustering by no k-means initializing')
self.means = gm.means_
self.variance = gm.covariances_
self.proportions = gm.weights_
return self.means, self.variance, self.proportions
def read_pred():
means = readFloat('%s-mixture_distribution_means.float3' %
predition_path) # shape (4, 2)
sigmas = readFloat('%s-mixture_distribution_sigmas.float3' %
predition_path) # shape (4, 2)
weights = readFloat('%s-mixture_distribution_weights.float3' %
predition_path) # shape (4)
sigmas = 2 * sigmas * sigmas
gmm = GaussianMixture(n_components=4, covariance_type='diag')
precisions_cholesky = _compute_precision_cholesky(sigmas, 'diag')
gmm.weights_ = weights
gmm.means_ = means
gmm.precisions_cholesky_ = precisions_cholesky
gmm.covariances_ = sigmas
return gmm
def EM_Process(data, n, covt):
'''
data: array shape data.
n: the number of components
covt: covariance_type
{‘full’, ‘tied’, ‘diag’, ‘spherical’}
chose one of them.
'''
GM = GaussianMixture(n_components=n,
covariance_type=covt,
max_iter=600,
random_state=3)
GM.means_ = np.zeros(3)
GM.covariances_ = np.identity(3)
GM.fit(data)
clusters = GM.predict(data)
return clusters
def train(self, train_data):
# 1. Create a GMM object and specify the number of components (classes) in the object
# 2. Fit the model to our training data. NOTE: You may need to reshape with np.reshape(-1,1)
# 3. Return None
data = np.array(train_data).reshape(-1, 1)
gmm = GaussianMixture(n_components=2)
fit = gmm.fit(data)
sort_indices = gmm.means_.argsort(axis=0)
order = sort_indices[:, 0]
gmm.means_ = gmm.means_[order, :]
gmm.covariances_ = gmm.covariances_[order, :]
w = np.split(gmm.weights_, 2)
w = np.asarray(w)
w = np.ravel(w[order, :])
gmm.weights_ = w
self.__model = gmm
return
def test_once_by_random_features():
Xtrain = numpy.random.random_sample((5000)).reshape(-1, 10)
Xtest = numpy.random.random_sample((500)).reshape(-1, 10)
gmm_orig = GaussianMixture(n_components=8, random_state=1)
gmm_copy = GaussianMixture()
gmm_orig.fit(Xtrain)
gmm_copy.weights_ = gmm_orig.weights_
gmm_copy.means_ = gmm_orig.means_
gmm_copy.covariances_ = gmm_orig.covariances_
gmm_copy.precisions_ = gmm_orig.precisions_
gmm_copy.precisions_cholesky_ = gmm_orig.precisions_cholesky_
gmm_copy.converged_ = gmm_orig.converged_
gmm_copy.n_iter_ = gmm_orig.n_iter_
gmm_copy.lower_bound_ = gmm_orig.lower_bound_
y_orig = gmm_orig.score_samples(Xtest)
y_copy = gmm_copy.score_samples(Xtest)
return all(y_orig == y_copy)
def test__estimate_log_prob_resp_spherical_shared_compression(self):
rs = np.random.RandomState(11)
cov_type = 'spherical'
gmm = GaussianMixture(n_components=3,
num_feat_full=5,
num_feat_comp=3,
num_feat_shared=3,
num_samp=4,
transform=None,
mask=None,
D_indices=None,
covariance_type=cov_type,
random_state=rs)
gmm.fit_sparsifier(X=self.td.X)
means = rs.rand(gmm.n_components, gmm.num_feat_full)
covariances = rs.rand(gmm.n_components)
weights = rs.rand(gmm.n_components)
weights /= weights.sum()
log_prob_test, log_resp_test, log_prob_norm_test = gmm._estimate_log_prob_resp(
weights, means, covariances, cov_type)
# find skl's values, pretty ugly to do.
precisions = _compute_precision_cholesky(covariances, cov_type)
gmm_skl = GMSKL(n_components=3, covariance_type=cov_type)
# we need the mask to be shared so that we can use mask[0] on all means
gmm_skl.means_ = means[:, gmm.mask[0]]
gmm_skl.precisions_cholesky_ = precisions
gmm_skl.weights_ = weights
gmm_skl.covariance_type_ = cov_type
log_prob_norm_true, log_resp_true = gmm_skl._estimate_log_prob_resp(
gmm.RHDX)
# if anything is bad later this overwrite with mean seems suspect:
log_prob_norm_true = log_prob_norm_true.mean()
# now get the log_prob from another function
log_prob_true = _estimate_log_gaussian_prob(gmm.RHDX, gmm_skl.means_,
precisions, cov_type)
# run the tests
self.assertArrayEqual(log_prob_test, log_prob_true)
self.assertArrayEqual(log_prob_norm_true, log_prob_norm_test)
self.assertArrayEqual(log_resp_true, log_resp_test)
def fit_markov_chain(y,plot=False): y_0 = y[:-1] y_1 = y[1:] grad_0 = np.gradient(y_0) grad_1 = np.gradient(y_1) state_1 = grad_1[np.where(grad_0 < 0)] # instances where previous gradient was negative state_2 = grad_1[np.where(grad_0 > 0)] # instances where previous gradient was positive mean_1,std_1 = stats.norm.fit(state_1) mean_2,std_2 = stats.norm.fit(state_2) # Reshaping parameters to be suitable for sklearn.GaussianMixture means = np.array([mean_1,mean_2]) means = means.reshape(2,1) y_GM = np.concatenate((state_2.reshape(-1,1),state_1.reshape(-1,1))) precisions = [1/(std_1**2),1/(std_2**2)] GM = GaussianMixture(n_components=2,covariance_type='spherical') GM.weights_ = [0.5,0.5] GM.means_ = means GM.covariances_ = [std_1,std_2] GM.precisions_ = precisions GM.precisions_cholesky_ = precisions GM.converged_ = True if(plot): samples = GM.sample(5000)[0] fig,ax_list = plt.subplots(3,1) fig.set_size_inches(20,20) ax_list[0].hist(state_1,bins=70) ax_list[1].hist(state_2,bins=70) lnspc_1 = np.linspace(state_1.min(),state_1.max(),y.shape[0]) gauss_1 = stats.norm.pdf(lnspc_1, mean_1, std_1) lnspc_2 = np.linspace(state_2.min(),state_2.max(),y.shape[0]) gauss_2 = stats.norm.pdf(lnspc_2, mean_2, std_2) ax_list[0].plot(lnspc_1,gauss_1) ax_list[1].plot(lnspc_2,gauss_2) ax_list[0].scatter(mean_1,30) ax_list[1].scatter(mean_2,30) ax_list[2].hist(samples,bins=100) plt.show() return GM
def fit_gmm_to_points(points,
n_components,
mdl,
ps=[],
num_iter=100,
covariance_type='full',
min_covar=0.001,
init_centers=[],
force_radii=-1.0,
force_weight=-1.0,
mass_multiplier=1.0):
"""fit a GMM to some points. Will return the score and the Akaike score.
Akaike information criterion for the current model fit. It is a measure
of the relative quality of the GMM that takes into account the
parsimony and the goodness of the fit.
if no particles are provided, they will be created
points: list of coordinates (python)
n_components: number of gaussians to create
mdl: IMP Model
ps: list of particles to be decorated. if empty, will add
num_iter: number of EM iterations
covariance_type: covar type for the gaussians. options: 'full', 'diagonal', 'spherical'
min_covar: assign a minimum value to covariance term. That is used to have more spherical
shaped gaussians
init_centers: initial coordinates of the GMM
force_radii: fix the radii (spheres only)
force_weight: fix the weights
mass_multiplier: multiply the weights of all the gaussians by this value
dirichlet: use the DGMM fitting (can reduce number of components, takes longer)
"""
new_sklearn = False
try:
from sklearn.mixture import GMM
except ImportError:
from sklearn.mixture import GaussianMixture
new_sklearn = True
print('creating GMM with n_components',n_components,'n_iter',num_iter,'covar type',covariance_type)
if new_sklearn:
# aic() calls size() on points, so it needs to a numpy array, not a list
points = np.array(points)
weights_init = precisions_init = None
if force_radii != -1.0:
print('warning: radii can no longer be forced, but setting '
'initial values to ', force_radii)
precisions_init = np.array([[1./force_radii]*3
for i in range(n_components)])
if force_weight != -1.0:
print('warning: weights can no longer be forced, but setting '
'initial values to ', force_weight)
weights_init = np.array([force_weight]*n_components)
gmm = GaussianMixture(n_components=n_components,
max_iter=num_iter,
covariance_type=covariance_type,
weights_init=weights_init,
precisions_init=precisions_init,
means_init=None if init_centers==[]
else init_centers)
else:
params='m'
init_params='m'
if force_radii==-1.0:
params+='c'
init_params+='c'
else:
covariance_type='spherical'
print('forcing spherical with radii',force_radii)
if force_weight==-1.0:
params+='w'
init_params+='w'
else:
print('forcing weights to be',force_weight)
gmm = GMM(n_components=n_components, n_iter=num_iter,
covariance_type=covariance_type, min_covar=min_covar,
params=params, init_params=init_params)
if force_weight!=-1.0:
gmm.weights_=np.array([force_weight]*n_components)
if force_radii!=-1.0:
gmm.covars_=np.array([[force_radii]*3 for i in range(n_components)])
if init_centers!=[]:
gmm.means_=init_centers
print('fitting')
model=gmm.fit(points)
score=gmm.score(points)
akaikescore=model.aic(points)
#print('>>> GMM score',gmm.score(points))
### convert format to core::Gaussian
if new_sklearn:
covars = gmm.covariances_
else:
covars = gmm.covars_
for ng in range(n_components):
covar=covars[ng]
if covar.size==3:
covar=np.diag(covar).tolist()
else:
covar=covar.tolist()
center=list(gmm.means_[ng])
weight=mass_multiplier*gmm.weights_[ng]
if ng>=len(ps):
ps.append(IMP.Particle(mdl))
shape=IMP.algebra.get_gaussian_from_covariance(covar,IMP.algebra.Vector3D(center))
g=IMP.core.Gaussian.setup_particle(ps[ng],shape)
IMP.atom.Mass.setup_particle(ps[ng],weight)
IMP.core.XYZR.setup_particle(ps[ng],sqrt(max(g.get_variances())))
return (score,akaikescore)
