Delay Embedding

Build delay-coordinate trajectories, lag matrices for autoregression, and sliding windows. Coordinates are returned as columns, with the newest sample first; rows advance in time.

Delay coordinates and autoregressive design matrices, using NumPy.

neuralfieldmanifold.embedding.delay(x, *, dimension, tau_samples)

Return [x(t), x(t-tau), ...] with shape (N-(d-1)*tau, d).

Rows are ordered in time; the first column is the newest sample. No centering, scaling, projection, or delay selection is performed.

neuralfieldmanifold.embedding.lag_matrix(x, *, order)

Return lag-1-first predictors and targets for an AR model.

neuralfieldmanifold.embedding.windows(x, *, length_samples, step_samples)

Return complete sliding windows and their integer start indices.