def calculate_log_likelihood(x, w, y):
latent_pred = x @ w
log_lik_elementwise = (y * norm.logcdf(latent_pred) + (1 - y) *
norm.logcdf(-latent_pred))
return np.sum(log_lik_elementwise)
def _log_1m_sf(self, params, T, Xs):
mu_params = params["mu_"]
mu_ = np.dot(Xs["mu_"], mu_params)
sigma_params = params["sigma_"]
log_sigma_ = np.dot(Xs["sigma_"], sigma_params)
sigma_ = np.exp(log_sigma_)
Z = (np.log(T) - mu_) / sigma_
return norm.logcdf(Z)
def _log_1m_sf(self, params, T, *Xs):
mu_params = params[self._LOOKUP_SLICE["mu_"]]
mu_ = np.dot(Xs[0], mu_params)
sigma_params = params[self._LOOKUP_SLICE["sigma_"]]
log_sigma_ = np.dot(Xs[1], sigma_params)
sigma_ = np.exp(log_sigma_)
Z = (np.log(T) - mu_) / sigma_
return norm.logcdf(Z, loc=0, scale=1)
def _log_1m_sf(self, params, T, *Xs):
mu_params = params[self._LOOKUP_SLICE["mu_"]]
mu_ = np.dot(Xs[0], mu_params)
sigma_params = params[self._LOOKUP_SLICE["sigma_"]]
log_sigma_ = np.dot(Xs[1], sigma_params)
sigma_ = np.exp(log_sigma_)
Z = (np.log(T) - mu_) / sigma_
return norm.logcdf(Z, loc=0, scale=1)
def _log_1m_sf(self, params, times):
mu_, ln_sigma_, lambda_ = params
sigma_ = safe_exp(ln_sigma_)
ilambda_2 = 1 / lambda_**2
Z = (log(times) - mu_) / sigma_
clipped_exp = np.clip(safe_exp(lambda_ * Z) * ilambda_2, 1e-15, 1e20)
if lambda_ > 0:
v = gammaincln(1 / lambda_**2, clipped_exp)
elif lambda_ < 0:
v = gammainccln(1 / lambda_**2, clipped_exp)
else:
v = norm.logcdf(Z)
return v
def logpx(x, mu=0.0, log_std=1.0, alpha=10.0):
return norm.logpdf(x, mu, np.exp(log_std)) + \
norm.logcdf(alpha * x, mu, np.exp(log_std)) + \
np.log(2)
def _log_1m_sf(self, params, times):
mu_, sigma_ = params
Z = (np.log(times) - mu_) / sigma_
return norm.logcdf(Z, loc=0, scale=1)
def probit_log_lik(fi, fj):
return norm.logcdf(fi - fj)
