def __repr__(self):
d = {
"01. Sampling module": self.sampling_module,
"02. Sampler": self.sampler,
"03. Number of samples": self.n_samples,
"04. Number of output parameters": self.n_outputs,
"05. Samples' shape": self.shape,
}
return formal_repr(self, d) + "\n06. Distribution parameters: " + dict_str(self.params) + \
"\n07 Resulting statistics: " + dict_str(self.stats)
def __repr__(self):
d = {
"01. Sampling module": self.sampling_module,
"02. Sampler": self.sampler,
"03. Number of samples": self.n_samples,
"04. Number of output parameters": self.n_outputs,
"05. Samples' shape": self.shape,
"06. Random seed": self.random_seed,
}
return formal_repr(self, d) + \
"\n07. Distribution parameters: " + dict_str(self.params) + \
"\n08. Truncation limits: " + str([dict_str(d) for d in dicts_of_lists_to_lists_of_dicts(self.trunc_limits)]) + \
"\n08. Resulting statistics: " + dict_str(self.stats)
def __repr__(self):
d = {
"01. Method": self.method,
"02. Second order calculation flag": self.calc_second_order,
"03. Confidence level": self.conf_level,
"05. Number of inputs": self.n_inputs,
"06. Number of outputs": self.n_outputs,
"07. Input names": self.input_names,
"08. Output names": self.output_names,
"09. Input bounds": self.input_bounds,
"10. Problem": dict_str(self.problem),
"11. Other parameters": dict_str(self.other_parameters),
}
return formal_repr(self, d)
mu = 0.25
std = 0.5
elif distribution == "beta":
mu = 0.5
std = 0.25
elif distribution == "binomial":
mu = 1.0
std = 1.0 / np.sqrt(2)
elif distribution == "chisquare":
mu = 1.0
std = np.sqrt(2 * mu)
else:
mu = 0.5
std = 0.5
logger.info(dict_str({"mu": mu, "std": std}))
p = mean_std_to_distribution_params(distribution, mu=mu, std=std)
logger.info(str(p))
logger.info("\nDistribution " + distribution + " to mu, std:")
mu1, std1 = distribution_params_to_mean_std(distribution, **p)
logger.info(dict_str({"mu": mu, "std": std}))
if np.abs(mu - mu1) > 10**-6 or np.abs(std - std1) > 10**-6:
raise ValueError("mu - mu1 = " + str(mu - mu1) + "std - std1 = " +
str(std - std1))
if distribution == "poisson":
mu = 0.25
std = 0.5
elif distribution == "beta":
mu = 0.5
std = 0.25
elif distribution == "binomial":
mu = 1.0
std = 1.0 / np.sqrt(2)
elif distribution == "chisquare":
mu = 1.0
std = np.sqrt(2 * mu)
else:
mu = 0.5
std = 0.5
LOG.info(dict_str({"mu": mu, "std": std}))
p = mean_std_to_distribution_params(distribution, mu=mu, std=std)
LOG.info(str(p))
LOG.info("\nDistribution " + distribution + " to mu, std:")
mu1, std1 = distribution_params_to_mean_std(distribution, **p)
LOG.info(dict_str({"mu": mu, "std": std}))
if np.abs(mu - mu1) > 10**-6 or np.abs(std - std1) > 10**-6:
raise ValueError("mu - mu1 = " + str(mu - mu1) + "std - std1 = " +
str(std - std1))