Method execute

physo/physym/program.py:285–306

Executes program on X. Parameters ---------- X : torch.tensor of shape (n_dim, ?,) of float Values of the input variables of the problem with n_dim = nb of input variables, ? = number of samples. i_realization : int, optional Index of

(self, X, i_realization = 0, n_samples_per_dataset = None)

        return y

    def execute(self, X, i_realization = 0, n_samples_per_dataset = None):
        """
        Executes program on X.
        Parameters
        ----------
        X : torch.tensor of shape (n_dim, ?,) of float
            Values of the input variables of the problem with n_dim = nb of input variables, ? = number of samples.
        i_realization : int, optional
            Index of realization to use for dataset specific free constants (0 by default).
        n_samples_per_dataset : array_like of shape (n_realizations,) of int or None, optional
            Overrides i_realization if given. If given assumes that X contains multiple datasets with samples of each
            dataset following each other and each portion of X corresponding to a dataset should be treated with its
            corresponding dataset specific free constants values. n_samples_per_dataset is the number of samples for
            each dataset. Eg. [90, 100, 110] for 3 datasets, this will assume that the first 90 samples of X are for
            the first dataset, the next 100 for the second and the last 110 for the third.
        Returns
        -------
        y : torch.tensor of shape (?,) of float
            Result of computation.
        """
        y = self.candidate_wrapper(lambda X: self.execute_wo_wrapper(X=X, i_realization=i_realization, n_samples_per_dataset=n_samples_per_dataset), X)
        return y

    def optimize_constants(self, X, y_target, y_weights = 1., i_realization = 0, n_samples_per_dataset = None, args_opti = None, freeze_class_free_consts = False):
        """