deeptime.covariance

The covariance package contains algorithms that can measure covariances for time series data.

Estimators

Covariance([lagtime, compute_c00, ...])	Compute (potentially lagged) covariances between data in an online fashion.
KoopmanWeightingEstimator(lagtime[, ...])	Computes Koopman operator and weights that can be plugged into the Covariance estimator.

Models

CovarianceModel([cov_00, cov_0t, cov_tt, ...])	A model which in particular carries the estimated covariances, means from a Covariance.
KoopmanWeightingModel(u, u_const, ...[, ...])	A model which contains the Koopman operator in a modified basis (PC|1) and can transform data into Koopman weights.

Utilities

WhiteningTransform(sqrt_inv_cov[, mean, dim])	Transformation of data into a whitened space.
covar(X[, remove_mean, modify_data, ...])	Computes the covariance matrix of X.
covars(X, Y[, remove_mean, modify_data, ...])	Computes the covariance and cross-covariance matrix of X and Y.
moments_XX(X[, remove_mean, modify_data, ...])	Computes the first two unnormalized moments of X.
moments_XXXY(X, Y[, remove_mean, ...])	Computes the first two unnormalized moments of X and Y.
moments_block(X, Y[, remove_mean, ...])	Computes the first two unnormalized moments of X and Y forward and backward.