maxfuse.model.Fusor.propagate

Fusor.propagate(wt1=0.7, wt2=0.7, svd_components1=None, svd_components2=None, metric='euclidean', randomized_svd=False, svd_runs=1, verbose=True)[source]

For indices not in pivots, find their matches by nearest neighbor search.

Parameters:

  • svd_components1 (None or int, default=None) – If not None, perform SVD to reduce the dimension of self.active_arr1 before doing internal nearest neighbor search.

  • svd_components2 (None or int, default=None) – If not None, perform SVD to reduce the dimension of self.active_arr1 before doing internal nearest neighbor search.

  • wt1 (float, default=0.7) – Weight to put on raw data of self.active_arr1 when doing smoothing.

  • wt2 (float, default=0.7) – Weight to put on raw data of self.active_arr2 when doing smoothing.

  • metric (string, default='correlation') – The metric to use in nearest neighbor search.

  • randomized_svd (bool, default=False) – Whether to perform randomized SVD.

  • svd_runs (int, default=1) – Perform multiple runs of SVD and the one with lowest Frobenious reconstruction error is selected.

  • verbose (bool, default=True) – Whether to print the progress.

Returns:

None