/ scikit-learn Public
Memory Error : Locally Linear Embedding #10493
Well-defined and straightforward way to resolve
Anything that requires some knowledge of conventions and best practices
Locally Linear Embedding (LLE) encounters a memory error when using a large matrix (10000x10000) and a large number of neighbors (>500).
The memory error occurs with the
standardoption after the nearest neighbors have been computed in
X[ind]creates a matrix of size
(n_samples x n_neighbors x n_dim)but the algorithm is then only using the neighbors sequentially in
barycenter_weights()could be modified to take
indas input and avoid this issue.
The text was updated successfully, but these errors were encountered: