Compute lambda_t = log(alpha_t) - log(sigma_t) of a given continuous-time label t in [0, T].
(self, t)
| 149 | return torch.sqrt(1. - torch.exp(2. * self.marginal_log_mean_coeff(t))) |
| 150 | |
| 151 | def marginal_lambda(self, t): |
| 152 | """ |
| 153 | Compute lambda_t = log(alpha_t) - log(sigma_t) of a given continuous-time label t in [0, T]. |
| 154 | """ |
| 155 | log_mean_coeff = self.marginal_log_mean_coeff(t) |
| 156 | log_std = 0.5 * torch.log(1. - torch.exp(2. * log_mean_coeff)) |
| 157 | return log_mean_coeff - log_std |
| 158 | |
| 159 | def inverse_lambda(self, lamb): |
| 160 | """ |
no test coverage detected