Alex Martens

+1

here is my work around: ``` command! -bang -nargs=? -complete=dir GFiles \ call fzf#vim#gitfiles(, {'options': ['-i']}, 0) noremap :GFiles --others --exclude-standard --cached ```

+1

I've got this problem too. I'm seeing `numpy.linalg.LinAlgError: LU decomposition error` ### Replication example * dataset: [my_data_sample.csv](https://drive.google.com/file/d/1hNdI65sRd_MmeK_FpS-Ec4zrQ4gAtW-O/view) * code: ``` import pandas as pd import statsmodels.api as sm endog =...

As per previous suggestion, ``` import numpy as np from scipy import linalg print(linalg.lapack.dgetrf([np.nan])) print(linalg.lapack.dgetrf([np.inf])) ``` Output: ``` (array([nan]), array([0], dtype=int32), 0) (array([inf]), array([0], dtype=int32), 0) ```

@ChadFulton, @bashtage, were you able to replicate this error?

I ran this model fitting on multiple `Ubuntu 18.04` systems with `Statsmodels 0.11.0.dev0+563.g044456ca7`. I just double checked and reran the code, and I'm still having the same error. Here is...

My results are the same as yours: ``` mod = sm.tsa.SARIMAX(np.zeros(10)) res = mod.filter([1, 1.]) print(res.filter_results.initial_state_cov) print(scipy.linalg.solve_discrete_lyapunov([[1 - 6e-17]], [[1.]])) print(scipy.linalg.solve_discrete_lyapunov([[1 - 5e-17]], [[1.]])) ``` ``` > [[-4.98960077e+291]] > [[4.50359963e+15]]...

> ```python > mod = sarimax.SARIMAX(np.zeros(10), order=(2, 0, 0)) > res = mod.filter([1, 2., 1.]) > ``` I see the LU decomposition error with this example too

I'm seeing the same problem on an M1 Mac. **OpenVPN Connect 3.3.6 (4368)**: Error calling protect () method on socket. Here's my log: ``` [Sep 15, 2022, 07:17:13] EVENT: RESOLVE...