def get_hdp_labels_gpu(gpu_data, w, mu, sigma, relabel=False):
labels = []
z = []
for i, data_set in enumerate(gpu_data):
densities = gpustats.mvnpdf_multi(data_set,
mu,
sigma,
weights=w[i].flatten(),
get=False,
logged=True,
order='C')
if relabel:
z.append(
np.asarray(
cuda_functions.gpu_apply_row_max(densities)[1].get(),
dtype='i'))
else:
z.append(None)
labels.append(
np.asarray(gpu_sampler.sample_discrete(densities, logged=True),
dtype='i'))
densities.gpudata.free()
del densities
return labels, z
def get_dp_labels_gpu(gpu_data, w, mu, sigma, relabel=False):
densities = gpustats.mvnpdf_multi(
gpu_data[0],
mu,
sigma,
weights=w.flatten(),
get=False,
logged=True,
order='C'
)
if relabel:
z = np.asarray(
cuda_functions.gpu_apply_row_max(densities)[1].get(),
dtype='i'
)
else:
z = None
labels = np.asarray(
gpu_sampler.sample_discrete(densities, logged=True),
dtype='i'
)
densities.gpudata.free()
del densities
return labels, z
def _compare_discrete(n, k):
densities = _make_test_densities(n, k)
dens = densities[0, :].copy() / densities[0, :].sum()
expected_mu = np.dot(np.arange(k), dens)
labels = gpusamp.sample_discrete(densities, logged=False)
est_mu = labels.mean()
return est_mu, expected_mu
def _compare_discrete(n, k):
densities = _make_test_densities(n, k)
dens = densities[0,:].copy() / densities[0,:].sum()
expected_mu = np.dot(np.arange(k), dens)
labels = gpusamp.sample_discrete(densities, logged=False)
est_mu = labels.mean()
return est_mu, expected_mu
def _compare_logged(n, k):
densities = np.log(_make_test_densities(n, k))
dens = np.exp((densities[0, :] - densities[0, :].max()))
dens = dens / dens.sum()
expected_mu = np.dot(np.arange(k), dens)
labels = gpusamp.sample_discrete(densities, logged=True)
est_mu = labels.mean()
return est_mu, expected_mu
def _compare_logged(n, k):
densities = np.log(_make_test_densities(n, k))
dens = np.exp((densities[0,:] - densities[0,:].max()))
dens = dens / dens.sum()
expected_mu = np.dot(np.arange(k), dens)
labels = gpusamp.sample_discrete(densities, logged=True)
est_mu = labels.mean()
return est_mu, expected_mu
def get_dp_labels_gpu(gpu_data, w, mu, sigma, relabel=False):
densities = gpustats.mvnpdf_multi(gpu_data[0],
mu,
sigma,
weights=w.flatten(),
get=False,
logged=True,
order='C')
if relabel:
z = np.asarray(cuda_functions.gpu_apply_row_max(densities)[1].get(),
dtype='i')
else:
z = None
labels = np.asarray(gpu_sampler.sample_discrete(densities, logged=True),
dtype='i')
densities.gpudata.free()
del densities
return labels, z
def get_hdp_labels_gpu(gpu_data, w, mu, sigma, relabel=False):
labels = []
z = []
for i, data_set in enumerate(gpu_data):
densities = gpustats.mvnpdf_multi(
data_set,
mu,
sigma,
weights=w[i].flatten(),
get=False,
logged=True,
order='C'
)
if relabel:
z.append(
np.asarray(
cuda_functions.gpu_apply_row_max(densities)[1].get(),
dtype='i'
)
)
else:
z.append(None)
labels.append(
np.asarray(
gpu_sampler.sample_discrete(densities, logged=True),
dtype='i'
)
)
densities.gpudata.free()
del densities
return labels, z
densities = gpustats.mvnpdf_multi(gdata[dataind], mu, Sigma,
weights=w.flatten(),
get=False, logged=True,
order='C')
if ttype == 2: # identification!!
Z = np.asarray(
cufuncs.gpu_apply_row_max(densities)[1].get(),
dtype='i'
)
else:
Z = None
labs = np.asarray(
gsamp.sample_discrete(densities, logged=True),
dtype='i'
)
subresult = [
np.array(nobs, dtype='i'),
labs,
np.array(gid, dtype='i')
]
if Z is not None:
subresult.append(Z)
results.append(subresult)
densities.gpudata.free()
del densities
#Sigma = Sigma.reshape(ncomp, ndim, ndim)
Sigma = a_Sigma[it]
if ttype>0: # 1 -- just densities ... 2 -- relabel too
## do GPU work ...
densities = gpustats.mvnpdf_multi(gdata[dataind], mu, Sigma,
weights = w.flatten(), get=False, logged=True,
order='C')
if ttype==2: #identification!!
Z = np.asarray(cufuncs.gpu_apply_row_max(densities)[1].get(), dtype='i')
else:
Z = None
labs = np.asarray(gsamp.sample_discrete(densities, logged=True), dtype='i')
subresult = [np.array(nobs, dtype='i'), labs, np.array(gid, dtype='i')]
if Z is not None:
subresult.append(Z)
results.append(subresult)
#subtask.labs = labs
#subtask.Z = Z
#subtask.nobs = nobs
#del subtask.mu
#del subtask.w
#del subtask.Sigma
densities.gpudata.free()
del densities
densities = gpustats.mvnpdf_multi(gdata[dataind],
mu,
Sigma,
weights=w.flatten(),
get=False,
logged=True,
order='C')
if ttype == 2: # identification!!
Z = np.asarray(
cufuncs.gpu_apply_row_max(densities)[1].get(),
dtype='i')
else:
Z = None
labs = np.asarray(gsamp.sample_discrete(densities,
logged=True),
dtype='i')
subresult = [
np.array(nobs, dtype='i'), labs,
np.array(gid, dtype='i')
]
if Z is not None:
subresult.append(Z)
results.append(subresult)
densities.gpudata.free()
del densities
elif ttype == 0:
# starting bem
