num_Ns = len(Ns)
num_gaussians = 5
plot_grid_size = 1000
# Number of trials to run
num_trials = 100
dists = np.zeros([num_trials,5,num_Ns])
for i, N in enumerate(Ns):
print 'Performing trials for N = %d'%N
for trial_num in range(num_trials):
# User feedback
print 'Running N = %d, trial_num = %d...'%(N, trial_num)
# Choose mixture of gaussians
gaussians = make_gaussian_mix(num_gaussians)
# Draw data from mixture
[xis, xgrid, Q_true, other] = draw_from_gaussian_mix(N=N, Nx=G, gaussians=gaussians)
# Compute data range and grid for fine-graned analysis
xmin = min(xgrid)
xmax = max(xgrid)
xint = [xmin, xmax]
xs = sp.linspace(xmin, xmax, plot_grid_size)
dx = xs[1]-xs[0]
# Interpolate Q_true for plotting
Q_true_vals = Q_true(xs)
# Perform DEFT density estimation
''' # Makes a simple figure demonstrating the inference proces. import scipy as sp from kinney2013_utils import make_gaussian_mix, draw_from_gaussian_mix, save_object from deft import deft_1d import time start_time = time.clock() # Design data distribution num_gaussians = 2 mus = [0, 10] sigmas = [2, 5] weights = [1, 1] gaussians = make_gaussian_mix(mus=mus, sigmas=sigmas, weights=weights) # Number of grid points to use for DEFT calculation G = 100 # Specify N and alphas N = 100 alpha = 2 num_samples = 20 # Draw data, rescaled to give an xint of length L = 10, centered on 0 xint = sp.array([-5.0, 5.0]) [xis, xint, Q_true, Q_true_details] = draw_from_gaussian_mix(N=N, Nx=G, gaussians=gaussians, xint=xint) xmid = sp.mean(xint) xspan = xint - xmid gaussians = Q_true_details.gaussians
num_Ns = len(Ns)
num_gaussians = 5
plot_grid_size = 1000
# Number of trials to run
num_trials = 100
dists = np.zeros([num_trials, 5, num_Ns])
for i, N in enumerate(Ns):
print 'Performing trials for N = %d' % N
for trial_num in range(num_trials):
# User feedback
print 'Running N = %d, trial_num = %d...' % (N, trial_num)
# Choose mixture of gaussians
gaussians = make_gaussian_mix(num_gaussians)
# Draw data from mixture
[xis, xgrid, Q_true,
other] = draw_from_gaussian_mix(N=N, Nx=G, gaussians=gaussians)
# Compute data range and grid for fine-graned analysis
xmin = min(xgrid)
xmax = max(xgrid)
xint = [xmin, xmax]
xs = sp.linspace(xmin, xmax, plot_grid_size)
dx = xs[1] - xs[0]
# Interpolate Q_true for plotting
Q_true_vals = Q_true(xs)
