MixedEffect doesn't work with numpy array - 'SVD did not converge'

[mtsotras]

• What is the current behavior?
  If using a numpy array to initialize a MixedEffect variable, the corresponding SLM model has a 'SVD did not converge' error when fitting to the data.

• Please provide the steps to reproduce and if possible a minimal demo of the problem.
  from brainstat.stats.terms import MixedEffect

term_subject = MixedEffect(demographics.SUB_ID.to_numpy(), 'SUB_ID')

model_mixed = term_age + term_sex + term_age * term_sex + term_subject

slm_mixed = SLM(
model_mixed,
-contrast_age,
surf=pial_combined,
mask=mask,
correction=["fdr", "rft"],
cluster_threshold=0.01,
two_tailed=False,
)
slm_mixed.fit(thickness)

• What is the expected behavior?

This code works if you do not convert the series to numpy array.

• Please tell us about your computing environment:
  Python 3.11.7

• Other information

[zihuaihuai]

Please start by checking and preprocessing your data to ensure there are no np.nan values, incorrect data types, or other issues in your NumPy array. Without access to your data, it's difficult to replicate the bug on our end.