def plotUpdateSummaryParallelFromDirectory(directory,
ax,
highlight=True,
plot_samples=False):
# Read data
costs = np.loadtxt(directory + "/costs.txt")
# Supress warnings http://stackoverflow.com/questions/19167550/prevent-or-dismiss-empty-file-warning-in-loadtxt
with warnings.catch_warnings():
warnings.simplefilter("ignore")
weights = np.loadtxt(directory + "/weights.txt")
if not weights:
# Sometimes weights.txt will be empty. Fill it with 1 in this case.
weights = np.empty(len(costs))
weights.fill(1)
n_parallel = np.loadtxt(directory + "/n_parallel.txt")
for i_parallel in range(n_parallel):
suffix = '_%02d' % i_parallel
distribution_mean = np.loadtxt(directory + "/distribution_mean" +
suffix + ".txt")
distribution_covar = np.loadtxt(directory + "/distribution_covar" +
suffix + ".txt")
samples = np.loadtxt(directory + "/samples" + suffix + ".txt")
distribution_new_mean = np.loadtxt(directory +
"/distribution_new_mean" + suffix +
".txt")
distribution_new_covar = np.loadtxt(directory +
"/distribution_new_covar" +
suffix + ".txt")
plotUpdateSummary(distribution_mean, distribution_covar, samples,
costs, weights, distribution_new_mean,
distribution_new_covar, ax, highlight, plot_samples)
def plotUpdateSummaryParallelFromDirectory(directory,ax,highlight=True,plot_samples=False):
# Read data
costs = np.loadtxt(directory+"/costs.txt")
weights = np.loadtxt(directory+"/weights.txt")
n_parallel = np.loadtxt(directory+"/n_parallel.txt")
for i_parallel in range(n_parallel):
suffix = '_%02d' % i_parallel
distribution_mean = np.loadtxt(directory+"/distribution_mean"+suffix+".txt")
distribution_covar = np.loadtxt(directory+"/distribution_covar"+suffix+".txt")
samples = np.loadtxt(directory+"/samples"+suffix+".txt")
distribution_new_mean = np.loadtxt(directory+"/distribution_new_mean"+suffix+".txt")
distribution_new_covar = np.loadtxt(directory+"/distribution_new_covar"+suffix+".txt")
plotUpdateSummary(distribution_mean,distribution_covar,samples,costs,weights,distribution_new_mean,distribution_new_covar,ax,highlight,plot_samples)
def plotUpdateSummaryParallelFromDirectory(directory,ax,highlight=True,plot_samples=False):
# Read data
costs = np.loadtxt(directory+"/costs.txt")
# Supress warnings http://stackoverflow.com/questions/19167550/prevent-or-dismiss-empty-file-warning-in-loadtxt
with warnings.catch_warnings():
warnings.simplefilter("ignore")
weights = np.loadtxt(directory+"/weights.txt")
if not weights:
# Sometimes weights.txt will be empty. Fill it with 1 in this case.
weights = np.empty(len(costs));
weights.fill(1)
n_parallel = np.loadtxt(directory+"/n_parallel.txt")
for i_parallel in range(n_parallel):
suffix = '_%02d' % i_parallel
distribution_mean = np.loadtxt(directory+"/distribution_mean"+suffix+".txt")
distribution_covar = np.loadtxt(directory+"/distribution_covar"+suffix+".txt")
samples = np.loadtxt(directory+"/samples"+suffix+".txt")
distribution_new_mean = np.loadtxt(directory+"/distribution_new_mean"+suffix+".txt")
distribution_new_covar = np.loadtxt(directory+"/distribution_new_covar"+suffix+".txt")
plotUpdateSummary(distribution_mean,distribution_covar,samples,costs,weights,distribution_new_mean,distribution_new_covar,ax,highlight,plot_samples)
def plotUpdateSummaryParallelFromDirectory(directory,
ax,
highlight=True,
plot_samples=False):
# Read data
costs = np.loadtxt(directory + "/costs.txt")
weights = np.loadtxt(directory + "/weights.txt")
n_parallel = np.loadtxt(directory + "/n_parallel.txt")
for i_parallel in range(n_parallel):
suffix = '_%02d' % i_parallel
distribution_mean = np.loadtxt(directory + "/distribution_mean" +
suffix + ".txt")
distribution_covar = np.loadtxt(directory + "/distribution_covar" +
suffix + ".txt")
samples = np.loadtxt(directory + "/samples" + suffix + ".txt")
distribution_new_mean = np.loadtxt(directory +
"/distribution_new_mean" + suffix +
".txt")
distribution_new_covar = np.loadtxt(directory +
"/distribution_new_covar" +
suffix + ".txt")
plotUpdateSummary(distribution_mean, distribution_covar, samples,
costs, weights, distribution_new_mean,
distribution_new_covar, ax, highlight, plot_samples)
