UMAP segmentation faults with correlation and cosine metrics in certain datasets

UMAP version 0.5.3
numba version 0.56.4

UMAP segfaults with correlation and cosine metrics but not with other metrics (euclidean, manhattan and camberra tested). The issues only occurs with very specific datasets.

Here is my code:

clusterable_embedding_test = umap.UMAP(
    n_neighbors=30,
    min_dist=.0,
    n_components=9,
    random_state=31416,
    #metric='manhattan',
    #metric='canberra',
    #metric='cosine',
    metric='correlation',
).fit_transform(df_clean)

I have tested in native Ubuntu 22.04, Ubuntu 22.04 under WSL2 and Windows. Same issue.

The issue only manifests itself when I remove a specific column from the dataset. If the column is present the code runs fine.

It does not matter how the column is removed (whether dropped from the dataframe, not loaded or removed beforehand from the CSV file).

Removing or adding other columns is OK, regardless of how many.

Setting NUMBA_DISABLE_JIT=1 prevents the segfault but results in extremely slow execution.

NUMBA_DISABLE_INTEL_SVML=1 has no effect. Same for a large number of other numba-related variables I have tried.

Start by reproducing the failure with UMAP 0.5.3 and numba 0.56.4 using the shown UMAP configuration, comparing correlation and cosine with the other tested metrics. Investigate why removing the specific dataset column triggers a native segfault and verify behavior across the reported environments. Done means the failing dataset no longer segfaults without disabling JIT, with a regression test or reproducible case.