if __name__ == '__main__':
np.random.seed(123)
J = 3
N = int(1e5)
K = 2
ncomps = 3
true_labels, data = generate_data(n=N, k=K, ncomps=ncomps)
data = data - data.mean(0)
data = data/data.std(0)
# shuffle the data...
ind = np.arange(N)
np.random.shuffle(ind)
all_data = data[ind].copy()
data = [all_data[(N/J*i):(N/J*(i+1))].copy() for i in range(J)]
mcmc = HDPNormalMixture(
data, ncomp=10, parallel=False, verbose=0, gpu=True
)
nburn = 20
niter = 2
iter_tracker = Callable(nburn, niter)
mcmc.sample(niter=niter, nburn=nburn, tune_interval=100, callback=iter_tracker)
print mcmc.mu[-1][-1]
J = 3
N = int(1e5)
K = 2
ncomps = 3
true_labels, data = generate_data(n=N, k=K, ncomps=ncomps)
data = data - data.mean(0)
data = data / data.std(0)
# shuffle the data...
ind = np.arange(N)
np.random.shuffle(ind)
all_data = data[ind].copy()
data = [all_data[(N / J * i):(N / J * (i + 1))].copy() for i in range(J)]
mcmc = HDPNormalMixture(data,
ncomp=10,
parallel=False,
verbose=0,
gpu=True)
nburn = 20
niter = 2
iter_tracker = Callable(nburn, niter)
mcmc.sample(niter=niter,
nburn=nburn,
tune_interval=100,
callback=iter_tracker)
print mcmc.mu[-1][-1]
from test_help import *
import numpy as np
from hdp import HDPNormalMixture
if __name__ == '__main__':
np.random.seed(123)
J = 3
N = int(1e5)
K = 2
ncomps = 3
true_labels, data = generate_data(n=N, k=K, ncomps=ncomps)
data = data - data.mean(0)
data = data/data.std(0)
# shuffle the data...
ind = np.arange(N)
np.random.shuffle(ind)
all_data = data[ind].copy()
data = [all_data[(N/J*i):(N/J*(i+1))].copy() for i in range(J)]
mcmc = HDPNormalMixture(
data, ncomp=10, parallel=False, verbose=1, gpu=True
)
mcmc.sample(niter=2, nburn=8, tune_interval=100)
print mcmc.mu[-1][-1]
#import gpustats as gs
if __name__ == '__main__':
J = 20
N = int(1e7)
K = 2
ncomps = 3
true_labels, data = generate_data(n=N, k=K, ncomps=ncomps)
data = data - data.mean(0)
data = data/data.std(0)
#shuffle the data ...
ind = np.arange(N); np.random.shuffle(ind);
all_data = data[ind].copy()
data = [ all_data[(N/J*i):(N/J*(i+1))].copy() for i in range(J) ]
#mcmc = HDPNormalMixture(data, ncomp=3, gpu=[0,1,2], parallel=True, verbose=100)
mcmc = HDPNormalMixture(data, ncomp=100, parallel=True, verbose=1,gpu=[0,1,2,3,4])
mcmc.sample(2, nburn=1, tune_interval=50)
#import pdb; pdb.set_trace()
#imcmc = HDPNormalMixture(mcmc, verbose=100)
#imcmc.sample(100, nburn=0, ident=True)
#print imcmc.mu[-1]
#print imcmc.weights[-1]
#print imcmc.beta[-1]
import numpy as np
from hdp import HDPNormalMixture
if __name__ == '__main__':
np.random.seed(123)
J = 3
N = int(1e5)
K = 2
ncomps = 3
true_labels, data = generate_data(n=N, k=K, ncomps=ncomps)
data = data - data.mean(0)
data = data / data.std(0)
# shuffle the data...
ind = np.arange(N)
np.random.shuffle(ind)
all_data = data[ind].copy()
data = [all_data[(N / J * i):(N / J * (i + 1))].copy() for i in range(J)]
mcmc = HDPNormalMixture(data,
ncomp=10,
parallel=False,
verbose=1,
gpu=True)
mcmc.sample(niter=2, nburn=8, tune_interval=100)
print mcmc.mu[-1][-1]
