I have a dataset with over 200.000 data samples with 256 features, then, I used UMAP with n_components = 8, 16, 32, 64, to reduce data dimension fron 256 to 64, 32, 16, 8, respectively. I do not have labels. I want to use umap validation embedding data. But I encountered an error "0xC00000FD" when I ran the umap validation.trustworthiness_vector(source=df_raw_data.to_numpy(), embedding=df_embedding.to_numpy(), max_k=K), K = 30. And Segmentation fault on WSL.How can I handle this situation?
I have tried to reduce n_components = 2, but the problem still happened.