Lmer.fixef no longer has group labels

[Shotgunosine]

1. Description

I'm pretty sure the fixef of LMER models used to have as its index the group labels. Now the index appears to be a range index.

2. Minimal reproducible example

In [1]: from pymer4.models import Lmer
...: from pymer4.utils import get_resource_path
...: import pandas as pd
...: import numpy as np
...: from scipy.special import logit
...: from scipy.stats import ttest_ind
...: import os

In [2]: import pymer4
...: pymer4.version
Out[2]: '0.7.2'

In [3]: df = pd.read_csv(os.path.join(get_resource_path(), "sample_data.csv"))
...: model = Lmer("DV ~ IV3 + IV2 + (IV2|Group) + (1|IV3)", data=df)
...: opt_opts = "optimizer='Nelder_Mead', optCtrl = list(FtolAbs=1e-8, XtolRel=1e-8)"
...: model.fit(summarize=False, control=opt_opts)
boundary (singular) fit: see ?isSingular

In [4]: model.fixef[0]
Out[4]:
(Intercept) IV3 IV2
0 6.031047 -1.529462 0.880139
1 20.356120 -1.529462 0.605304
2 10.322812 -1.529462 0.836379
3 12.222470 -1.529462 0.858911
4 11.251184 -1.529462 0.198474
5 12.626052 -1.529462 0.902386
6 2.466363 -1.529462 0.082339
7 -1.951063 -1.529462 1.070559
8 41.427467 -1.529462 0.256847
9 -0.327156 -1.529462 0.532686
10 3.409337 -1.529462 0.826673
11 4.183722 -1.529462 1.022796
12 17.441838 -1.529462 -0.170657
13 16.894473 -1.529462 0.719052
14 9.393190 -1.529462 0.630868
15 20.406180 -1.529462 0.112961
16 16.228585 -1.529462 0.670403
17 -0.947947 -1.529462 0.977892
18 2.960673 -1.529462 0.881681
19 15.081342 -1.529462 0.903035
20 3.099670 -1.529462 0.271275
21 11.046744 -1.529462 0.898630
22 1.995796 -1.529462 0.827309
23 0.502841 -1.529462 0.637432
24 16.325339 -1.529462 0.684816
25 28.728807 -1.529462 0.552281
26 0.282634 -1.529462 1.083250
27 12.274335 -1.529462 0.778643
28 2.900907 -1.529462 0.997698
29 25.969906 -1.529462 0.569460
30 18.774758 -1.529462 0.784060
31 41.014190 -1.529462 0.384460
32 17.825418 -1.529462 -0.101800
33 1.311976 -1.529462 0.714895
34 7.733580 -1.529462 0.970813
35 7.566232 -1.529462 0.603793
36 2.572444 -1.529462 0.991525
37 30.764099 -1.529462 0.557071
38 7.069487 -1.529462 0.870455
39 11.825799 -1.529462 0.858441
40 15.998152 -1.529462 0.395001
41 16.461600 -1.529462 0.839032
42 9.140041 -1.529462 0.900363
43 15.011776 -1.529462 0.765686
44 9.947849 -1.529462 0.858361
45 -2.077808 -1.529462 1.090571
46 32.480301 -1.529462 0.269124

In [5]: df.Group.unique()
Out[5]:
array([ 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 14, 15, 16, 17, 18, 19,
20, 21, 22, 23, 24, 27, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39,
40, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53])

In [6]: model.fixef[0].index
Out[6]: RangeIndex(start=0, stop=47, step=1)

3. Conda environment

Conda environment created with the following command
conda create -p potdys3 -c ejolly -c conda-forge python=3.8.8 numpy=1.19.2 pandas=1.1.5 pytest=6.2.2 joblib=1.0.1 rpy2=3.4.3 matplotlib=3.3.4 seaborn=0.11.1 scikit-learn=0.24.1 numexpr=2.7.3 patsy=0.5.1 statsmodels=0.12.2 openpyxl=3.0.7 pymer4=0.7.2 ipython jupyter notebook

conda list --explicit > pymer4_issue.txt
[pymer4_issue.txt](https://github.com/ejolly/pymer4/files/6286562/pymer4_issue.txt)

4. System information

• Mac OSX, in a terminal window

	$ uname -mprsv
	Darwin 18.7.0 Darwin Kernel Version 18.7.0: Tue Jan 12 22:04:47 PST 2021; root:xnu-4903.278.56~1/RELEASE_X86_64 x86_64 i386

	$ sw_vers
        ProductName:	Mac OS X
        ProductVersion:	10.14.6
        BuildVersion:	18G8022

[Shotgunosine]

Just confirmed that as of version 0.7.0 the index of the fixef was the group labels.