pflm.fpca.utils.get_fpca_ce_score#

get_fpca_ce_score(flatten_func_data: FlattenFunctionalData, mu: ndarray, num_pcs: int, fpca_lambda: ndarray, fpca_phi: ndarray, fitted_cov: ndarray, sigma2: float) → tuple[ndarray, list[ndarray], list[ndarray]][source][source]#

Compute conditional expectation (CE) FPCA scores and fitted curves.

Parameters:

flatten_func_dataFlattenFunctionalData: Flattened data containing fields y, t, tid, unique_sid, sid_cnt.
munp.ndarray of shape (nt,): Mean on the observation grid.
num_pcsint: Number of principal components to use (<= len(fpca_lambda)).
fpca_lambdanp.ndarray of shape (k,): Eigenvalues for FPCA components.
fpca_phinp.ndarray of shape (nt, k): Basis functions on the observation grid (columns are components).
fitted_covnp.ndarray of shape (nt, nt): Fitted covariance on the observation grid.
sigma2float: Measurement noise variance.

Returns:

xinp.ndarray of shape (n_samples, num_pcs): CE scores by subject.
xi_varList[np.ndarray]: Per-subject score covariance matrices or variance summaries.
fitted_y_matnp.ndarray of shape (nt, n_samples): Fitted values on the observation grid.
fitted_yList[np.ndarray]: Fitted values at the observed time points per subject.

Raises:

ValueError: If fitted_cov has wrong shape, num_pcs exceeds available eigenvalues, or fpca_phi has incompatible shape.