This PR aims to solve https://github.com/bambinos/bambi/issues/439.

I have pushed the first version of the models. Here are the formulas I used:

• pooled_model: log_radon ~ floor
• unpooled_model: log_radon ~ 0 + county + county:floor
• partial_pooling_model: log_radon ~ 0 + (0 + 1|county)
• varying_intercept_model: log_radon ~ 0 + (1|county) + floor # ! How to do complete one-hot-encoding for county?
• varying_intercept_slope_model: log_radon ~ 0 + (floor|county) # ! How to do complete one-hot-encoding for county?

I could like to have all the countries as encoded (not removing the first one) plus I am not auto-scaling the data to make the results with the original PyMC example comparable. Does this look good? I am still getting used to the formula-like notation for hierarchical models (which is why I am working on this PR 😅 ).

After double-checking the model specification I will continue to add text and some additional plots.

This PR adds a new notebook where Bambi is used to fit a couple of logistic regression models and then Arviz is used to perform model comparison. I would like to have feedback on the example before merging.

Note: This was mentioned in #210

I updated some notebooks to reflect changes in Bambi and some extra comments/explanations I added when considered they could be helpful. I include the WIP label because I think the notebooks may be modified before merging according to feedback and discussion.

Note: In a chat with @aloctavodia we talked about modifying the object that is passed to az.plot_trace() in such a way that *_offset terms are not included by default. This is not included in this PR. I think we should work on it in a separate PR and then update the examples again to reflect changes.

When a model with default priors and non-identifiable fixed effects is created (e.g., by calling something like add_term('condition', drop_first=False) when an intercept is already in the model), the prior scaling approach currently implemented in PriorScaler sets the random factor hyperprior variances to inf, and the sampler consequently refuses to start. We should either check for this scenario and substitute some large but sane value, or raise an exception with a more informative message rather than waiting for Theano to complain.

This PR adds a new StanBackEnd that should (in theory) do almost everything that the PyMC3BackEnd does. Using it is as simple as specifying backend='stan' when initializing a new Model. The StanBackEnd wraps PyStan, but converts the sampling output into a PyMC3 MultiTrace object, which allows us to use all of the existing plotting/summarizing machinery (at least in principle; in practice, I had to hack my way around the normal MultiTrace initialization process, so some attributes are probably not set properly).

There are some things to iron out before we can merge this. @jake-westfall, maybe you can take a pass at these? We need to do the following:

• Test everything more thoroughly to make sure it works.

• Modify at least a few of the existing tests to run using the Stan backend. It's probably also worth adding a couple of tests that run with both backends and make sure the resulting estimates are close. We also need to modify the travis config to install PyStan (and maybe add an optional_dependencies.txt file containing PyStan that people can pip install -r from).

• Add support for all of the PyMC3 distributions we currently support. The trick here is to map PyMC3 distribution names and arguments onto the Stan language. This is accomplished via the dists dictionary in StanBackEnd, which maps the PyMC3 dist names onto dictionaries containing the Stan names, argument order, and (optionally) and bounds to impose. The latter is required because in Stan there are no separate distributions for clipped distributions; e.g., half-cauchy is just cauchy with a lower bound of 0. In principle I think all we need to do is add new entries to the dists dictionary; I don't think any of the PyMC3 distributions we currently support should introduce any new complications.

• One gotcha right now is that PyStan ignores the incl_warmup argument in extract() if permuted=True. This is reasonable in one sense, because the permuted results concatenate all chains together, and it's arguably weird to include the different initialization points. But it's a problem for us, because to make results commensurable with the PyMC3BackEnd, we should really include the burn-in and let the user drop it themselves later. The solution is unfortunately probably going to require us to duplicate some of the PyStan code for reformatting the Stan output, but with the burn-in included.

• Add support for different error distributions. Right now the StanBackEnd only supports normally distributed errors, but we should at minimum support the options we support with PyMC3. This shouldn't take much work; I think we just need to modify the specification of the yhat term and the model error.

Thanks for all the great work with Bambi. It has been a godsend for my work.

I tried to implement the following formula:

y ~ time * tx + (time | therapist/subjects)

which expands to:

y ~ time * tx + (time | therapist:subjects) + (time | therapist)

and I get the following error:

"ValueError: Factor on right hand side of group specific term must be a single term"

Is it possible to implement three-level conditional hierarchical models or am I simply writing this incorrectly? If I am writing it incorrectly, how would I need to reformat it to work with bambi/formulae?

This is to address https://github.com/bambinos/bambi/issues/522 and https://github.com/bambinos/bambi/issues/525 inspired by @zwelitunyiswa's example

I decided to add a single new value to the fit() method which allows switching in of numpyro/blackjax samplers instead of the pymc default. I decided against some cpu/gpu flags because it's mostly decided by whatever Jax can find and the methods I saw to disable GPUs are quite hacky involving playing with your environment variables which I felt is out of scope for a library to be fiddling with so I've just noted this in the documentation instead.

I've tested the samplers locally and they work on one of my personal projects, but I'll try and knock up a simple example shortly which demonstrates them all.

One note: The PyMC 4 release blog post says:

These samplers live in a different submodule sampling_jax but the plan is to integrate them into pymc.sample(backend="JAX").

So we should expect the implementation here to change pretty soon, so I think it's worth keeping the implementation in bambi simple so it's easy to port-over when this happens.

Fixes #81

Notes on reviewing

• The diff count shows that this PR is large, but its quite misleading. A lot of the line count is either from
  • autogenerated files (conf.py, Make.bat, and Makefile) -- out of these one should only review conf.py if you're familiar with sphinx
  • ipython notebook diffs. I only updated slight wording and the number of octothorpes (#) in the headers, did not re-run any cells
• all code in bambi/ are docstring related. I did not add or remove any documentation, but there were some spacing problems (which cause sphinx to have a hard time)
  • the only change of note here is that I changed some docstrings in bambi/diagnostics.py from the numpy convention to the Google convention, since most of the library uses the Google convention
• I would pretty much only look at the .rst file changes
• Happy to hear any thoughts people have whether it be reversions, continued code changes, or to back up any decisions I made!

Changes

• Incorporates more information from the README.md
• Adds example notebooks
  • Made the examples/ directory a symbolic link to the notebooks which are now housed in the documentation (similar to PyMC3)
  • Edited the headers in the notebooks since they are read in by sphinx (mostly this is making one top-level header # and making the other headers either ## or ###
  • No other edits to notebooks were made (mention this since notebooks are hard to review)
• API Reference now includes all classes, methods and functions from models/, priors/, results/, and diagnostics.

To run the docs locally

0. Grab this branch, and enter the root directory of this repository
1. enter an python3 virtual env of your choice (I used conda create python=3.6)
2. install the library python setup.py install
3. install requirements pip install -r requirements-dev.txt
4. enter the docs directory cd docs/
5. run the build make html
6. open the generated html in browser open _build/html/index.html (from still within the docs/ directory

Next steps

• hosting on readthedocs?
• If there is any custom CSS or images that people want on the docs?

Added a notebook with examples using the categorical family to address #436. I noticed on the original merge (#426) that @tomicapretto made several very nice examples, so I aggregated several into one notebook and added some comments to tie it together.

Note that I omitted two of the examples, the satisfaction survey and the inhaler example. I made that choice because I think those two fall more in the domain of ordinal regression, rather than categorical. I'm happy to add them if you'd prefer to include them though.

I'm working on a laptop with Windows 10 and using Anaconda 64-bit. I have created an environment for working with Pymc3 and bambi. I have been able to test Pymc3 and it is working for an Hierarchical Linear Regression model. For bambi I was testing the example described at this link (growth curves of Pigs example) [https://bambinos.github.io/bambi/master/notebooks/multi-level_regression.html]. I run into an error I had not seen with bambi before (actually from the output this looks like it is occurring at Theano, but I'm not a programmer or a developer).

Here is my environment and the complete traceback of the error

# packages in environment at C:\ProgramData\Anaconda3\envs\pm3env:
#
# Name                    Version                   Build  Channel
argon2-cffi               20.1.0           py37hcc03f2d_2    conda-forge
arviz                     0.11.1                   pypi_0    pypi
async_generator           1.10                       py_0    conda-forge
attrs                     21.2.0             pyhd8ed1ab_0    conda-forge
backcall                  0.2.0              pyh9f0ad1d_0    conda-forge
backports                 1.0                        py_2    conda-forge
backports.functools_lru_cache 1.6.4              pyhd8ed1ab_0    conda-forge
bambi                     0.5.0                    pypi_0    pypi
blas                      1.0                         mkl
bleach                    4.0.0              pyhd8ed1ab_0    conda-forge
ca-certificates           2021.5.30            h5b45459_0    conda-forge
cached-property           1.5.2                    pypi_0    pypi
cachetools                4.2.2                    pypi_0    pypi
cairo                     1.16.0            hb19e0ff_1008    conda-forge
certifi                   2021.5.30        py37h03978a9_0    conda-forge
cffi                      1.14.6           py37hd8e9650_0    conda-forge
cftime                    1.5.0                    pypi_0    pypi
colorama                  0.4.4              pyh9f0ad1d_0    conda-forge
cycler                    0.10.0                     py_2    conda-forge
decorator                 5.0.9              pyhd8ed1ab_0    conda-forge
defusedxml                0.7.1              pyhd8ed1ab_0    conda-forge
dill                      0.3.4                    pypi_0    pypi
entrypoints               0.3             pyhd8ed1ab_1003    conda-forge
et-xmlfile                1.1.0                    pypi_0    pypi
expat                     2.4.1                h39d44d4_0    conda-forge
fastprogress              1.0.0                    pypi_0    pypi
filelock                  3.0.12                   pypi_0    pypi
font-ttf-dejavu-sans-mono 2.37                 hab24e00_0    conda-forge
font-ttf-inconsolata      3.000                h77eed37_0    conda-forge
font-ttf-source-code-pro  2.038                h77eed37_0    conda-forge
font-ttf-ubuntu           0.83                 hab24e00_0    conda-forge
fontconfig                2.13.1            h1989441_1005    conda-forge
fonts-conda-ecosystem     1                             0    conda-forge
fonts-conda-forge         1                             0    conda-forge
formulae                  0.1.3                    pypi_0    pypi
freetype                  2.10.4               h546665d_1    conda-forge
fribidi                   1.0.10               h8d14728_0    conda-forge
getopt-win32              0.1                  h8ffe710_0    conda-forge
gettext                   0.19.8.1          h1a89ca6_1005    conda-forge
graphite2                 1.3.13                     1000    conda-forge
graphviz                  2.48.0               hefbd956_0    conda-forge
gts                       0.7.6                h7c369d9_2    conda-forge
h5py                      3.1.0                    pypi_0    pypi
harfbuzz                  2.8.2                hc601d6f_0    conda-forge
icc_rt                    2019.0.0             h0cc432a_1
icu                       68.1                 h0e60522_0    conda-forge
importlib-metadata        2.1.1                    pypi_0    pypi
intel-openmp              2021.3.0          h57928b3_3372    conda-forge
ipykernel                 5.5.5            py37h7813e69_0    conda-forge
ipython                   7.26.0           py37h4038f58_0    conda-forge
ipython_genutils          0.2.0                      py_1    conda-forge
ipywidgets                7.6.3              pyhd3eb1b0_1
jbig                      2.1               h8d14728_2003    conda-forge
jedi                      0.18.0           py37h03978a9_2    conda-forge
jinja2                    3.0.1              pyhd8ed1ab_0    conda-forge
jpeg                      9d                   h8ffe710_0    conda-forge
jsonschema                3.2.0              pyhd8ed1ab_3    conda-forge
jupyter                   1.0.0                    py37_7
jupyter_client            6.1.12             pyhd8ed1ab_0    conda-forge
jupyter_console           6.4.0              pyhd8ed1ab_0    conda-forge
jupyter_core              4.7.1            py37h03978a9_0    conda-forge
jupyterlab_pygments       0.1.2              pyh9f0ad1d_0    conda-forge
jupyterlab_widgets        1.0.0              pyhd8ed1ab_1    conda-forge
kiwisolver                1.3.1            py37h8c56517_1    conda-forge
lcms2                     2.12                 h2a16943_0    conda-forge
lerc                      2.2.1                h0e60522_0    conda-forge
libclang                  11.1.0          default_h5c34c98_1    conda-forge
libdeflate                1.7                  h8ffe710_5    conda-forge
libffi                    3.3                  h0e60522_2    conda-forge
libgd                     2.3.2                h138e682_0    conda-forge
libglib                   2.68.3               h1e62bf3_0    conda-forge
libiconv                  1.16                 he774522_0    conda-forge
libpng                    1.6.37               h1d00b33_2    conda-forge
libpython                 2.1                      py37_0
libsodium                 1.0.18               h8d14728_1    conda-forge
libtiff                   4.3.0                h0c97f57_1    conda-forge
libwebp                   1.2.0                h57928b3_0    conda-forge
libwebp-base              1.2.0                h8ffe710_2    conda-forge
libxcb                    1.13              hcd874cb_1003    conda-forge
libxml2                   2.9.12               hf5bbc77_0    conda-forge
llvmlite                  0.36.0           py37habb0c8c_0    conda-forge
lz4-c                     1.9.3                h8ffe710_1    conda-forge
m2w64-gcc-libgfortran     5.3.0                         6    conda-forge
m2w64-gcc-libs            5.3.0                         7    conda-forge
m2w64-gcc-libs-core       5.3.0                         7    conda-forge
m2w64-gmp                 6.1.0                         2    conda-forge
m2w64-libwinpthread-git   5.0.0.4634.697f757               2    conda-forge
markupsafe                2.0.1            py37hcc03f2d_0    conda-forge
matplotlib                3.3.2                haa95532_0
matplotlib-base           3.3.2            py37h3379fd5_1    conda-forge
matplotlib-inline         0.1.2              pyhd8ed1ab_2    conda-forge
mistune                   0.8.4           py37hcc03f2d_1004    conda-forge
mkl                       2020.4             hb70f87d_311    conda-forge
mkl-service               2.3.0            py37h196d8e1_0
mkl_fft                   1.3.0            py37hda49f71_1    conda-forge
mkl_random                1.2.0            py37h414f9d2_1    conda-forge
mpmath                    1.2.1              pyhd8ed1ab_0    conda-forge
msys2-conda-epoch         20160418                      1    conda-forge
nbclient                  0.5.3              pyhd8ed1ab_0    conda-forge
nbconvert                 6.1.0            py37h03978a9_0    conda-forge
nbformat                  5.1.3              pyhd8ed1ab_0    conda-forge
nest-asyncio              1.5.1              pyhd8ed1ab_0    conda-forge
netcdf4                   1.5.7                    pypi_0    pypi
notebook                  6.4.0              pyha770c72_0    conda-forge
numba                     0.53.1           py37h4e635f9_0    conda-forge
numpy                     1.19.2           py37hadc3359_0
numpy-base                1.19.2           py37ha3acd2a_0
olefile                   0.46               pyh9f0ad1d_1    conda-forge
openjpeg                  2.4.0                hb211442_1    conda-forge
openpyxl                  3.0.7                    pypi_0    pypi
openssl                   1.1.1k               h8ffe710_0    conda-forge
packaging                 21.0               pyhd8ed1ab_0    conda-forge
pandas                    1.2.1            py37hf11a4ad_0
pandoc                    2.14.1               h8ffe710_0    conda-forge
pandocfilters             1.4.2                      py_1    conda-forge
pango                     1.48.7               hd84fcdd_0    conda-forge
parso                     0.8.2              pyhd8ed1ab_0    conda-forge
patsy                     0.5.1                    pypi_0    pypi
pcre                      8.45                 h0e60522_0    conda-forge
pickleshare               0.7.5                   py_1003    conda-forge
pillow                    8.3.1            py37hd7d9ad0_0    conda-forge
pip                       20.3.3           py37haa95532_0
pixman                    0.40.0               h8ffe710_0    conda-forge
prometheus_client         0.11.0             pyhd8ed1ab_0    conda-forge
prompt-toolkit            3.0.19             pyha770c72_0    conda-forge
prompt_toolkit            3.0.19               hd8ed1ab_0    conda-forge
pthread-stubs             0.4               hcd874cb_1001    conda-forge
pycparser                 2.20               pyh9f0ad1d_2    conda-forge
pygments                  2.9.0              pyhd8ed1ab_0    conda-forge
pymc3                     3.11.2                   pypi_0    pypi
pyparsing                 2.4.7              pyh9f0ad1d_0    conda-forge
pyqt                      5.12.3           py37h03978a9_7    conda-forge
pyqt-impl                 5.12.3           py37hf2a7229_7    conda-forge
pyqt5-sip                 4.19.18          py37hf2a7229_7    conda-forge
pyqtchart                 5.12             py37hf2a7229_7    conda-forge
pyqtwebengine             5.12.1           py37hf2a7229_7    conda-forge
pyrsistent                0.17.3           py37hcc03f2d_2    conda-forge
python                    3.7.9                h60c2a47_0
python-dateutil           2.8.2              pyhd8ed1ab_0    conda-forge
python-graphviz           0.16               pyh243d235_2    conda-forge
python_abi                3.7                     2_cp37m    conda-forge
pytz                      2021.1             pyhd8ed1ab_0    conda-forge
pywin32                   300              py37hcc03f2d_0    conda-forge
pywinpty                  1.1.3            py37h7f67f24_0    conda-forge
pyzmq                     22.2.0           py37hcce574b_0    conda-forge
qt                        5.12.9               h5909a2a_4    conda-forge
qtconsole                 5.1.1              pyhd8ed1ab_0    conda-forge
qtpy                      1.9.0                      py_0    conda-forge
scipy                     1.7.1                    pypi_0    pypi
semver                    2.13.0                   pypi_0    pypi
send2trash                1.7.1              pyhd8ed1ab_0    conda-forge
setuptools                49.6.0           py37h03978a9_3    conda-forge
six                       1.16.0             pyh6c4a22f_0    conda-forge
sqlite                    3.36.0               h8ffe710_0    conda-forge
statsmodels               0.12.2                   pypi_0    pypi
sympy                     1.7.1            py37h03978a9_1    conda-forge
terminado                 0.10.1           py37h03978a9_0    conda-forge
testpath                  0.5.0              pyhd8ed1ab_0    conda-forge
theano-pymc               1.1.2                    pypi_0    pypi
tk                        8.6.10               h8ffe710_1    conda-forge
tornado                   6.1              py37hcc03f2d_1    conda-forge
traitlets                 5.0.5                      py_0    conda-forge
typing                    3.7.4.3                  pypi_0    pypi
typing_extensions         3.10.0.0           pyha770c72_0    conda-forge
ucrt                      10.0.20348.0         h57928b3_0    conda-forge
vc                        14.2                 hb210afc_5    conda-forge
vs2015_runtime            14.29.30037          h902a5da_5    conda-forge
watermark                 2.2.0                    pypi_0    pypi
wcwidth                   0.2.5              pyh9f0ad1d_2    conda-forge
webencodings              0.5.1                      py_1    conda-forge
wheel                     0.36.2             pyhd3deb0d_0    conda-forge
widgetsnbextension        3.5.1            py37h03978a9_4    conda-forge
wincertstore              0.2             py37h03978a9_1006    conda-forge
winpty                    0.4.3                         4    conda-forge
xarray                    0.16.2                   pypi_0    pypi
xorg-kbproto              1.0.7             hcd874cb_1002    conda-forge
xorg-libice               1.0.10               hcd874cb_0    conda-forge
xorg-libsm                1.2.3             hcd874cb_1000    conda-forge
xorg-libx11               1.7.2                hcd874cb_0    conda-forge
xorg-libxau               1.0.9                hcd874cb_0    conda-forge
xorg-libxdmcp             1.1.3                hcd874cb_0    conda-forge
xorg-libxext              1.3.4                hcd874cb_1    conda-forge
xorg-libxpm               3.5.13               hcd874cb_0    conda-forge
xorg-libxt                1.2.1                hcd874cb_2    conda-forge
xorg-xextproto            7.3.0             hcd874cb_1002    conda-forge
xorg-xproto               7.0.31            hcd874cb_1007    conda-forge
xz                        5.2.5                h62dcd97_1    conda-forge
zeromq                    4.3.4                h0e60522_0    conda-forge
zipp                      3.5.0              pyhd8ed1ab_0    conda-forge
zlib                      1.2.11            h62dcd97_1010    conda-forge
zstd                      1.5.0                h6255e5f_0    conda-forge

Model I tried to run is:

model = bmb.Model("Weight ~ Time + (Time|Pig)", data)
results = model.fit()

And the error I get is:

---------------------------------------------------------------------------
Exception                                 Traceback (most recent call last)
<ipython-input-11-99071b9bde96> in <module>
      1 model = bmb.Model("Weight ~ Time + (Time|Pig)", data)
----> 2 results = model.fit()

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\bambi\models.py in fit(self, omit_offsets, backend, **kwargs)
    213             )
    214 
--> 215         return self.backend.run(omit_offsets=omit_offsets, **kwargs)
    216 
    217     def build(self, backend="pymc"):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\bambi\backends\pymc.py in run(self, start, method, init, n_init, omit_offsets, **kwargs)
    139                     n_init=n_init,
    140                     return_inferencedata=True,
--> 141                     **kwargs,
    142                 )
    143 

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\pymc3\sampling.py in sample(draws, step, init, n_init, start, trace, chain_idx, chains, cores, tune, progressbar, model, random_seed, discard_tuned_samples, compute_convergence_checks, callback, jitter_max_retries, return_inferencedata, idata_kwargs, mp_ctx, pickle_backend, **kwargs)
    502                 progressbar=progressbar,
    503                 jitter_max_retries=jitter_max_retries,
--> 504                 **kwargs,
    505             )
    506             if start is None:

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\pymc3\sampling.py in init_nuts(init, chains, n_init, model, random_seed, progressbar, jitter_max_retries, **kwargs)
   2185         raise ValueError(f"Unknown initializer: {init}.")
   2186 
-> 2187     step = pm.NUTS(potential=potential, model=model, **kwargs)
   2188 
   2189     return start, step

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\pymc3\step_methods\hmc\nuts.py in __init__(self, vars, max_treedepth, early_max_treedepth, **kwargs)
    166         `pm.sample` to the desired number of tuning steps.
    167         """
--> 168         super().__init__(vars, **kwargs)
    169 
    170         self.max_treedepth = max_treedepth

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\pymc3\step_methods\hmc\base_hmc.py in __init__(self, vars, scaling, step_scale, is_cov, model, blocked, potential, dtype, Emax, target_accept, gamma, k, t0, adapt_step_size, step_rand, **theano_kwargs)
     86         vars = inputvars(vars)
     87 
---> 88         super().__init__(vars, blocked=blocked, model=model, dtype=dtype, **theano_kwargs)
     89 
     90         self.adapt_step_size = adapt_step_size

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\pymc3\step_methods\arraystep.py in __init__(self, vars, model, blocked, dtype, logp_dlogp_func, **theano_kwargs)
    252 
    253         if logp_dlogp_func is None:
--> 254             func = model.logp_dlogp_function(vars, dtype=dtype, **theano_kwargs)
    255         else:
    256             func = logp_dlogp_func

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\pymc3\model.py in logp_dlogp_function(self, grad_vars, tempered, **kwargs)
   1002         varnames = [var.name for var in grad_vars]
   1003         extra_vars = [var for var in self.free_RVs if var.name not in varnames]
-> 1004         return ValueGradFunction(costs, grad_vars, extra_vars, **kwargs)
   1005 
   1006     @property

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\pymc3\model.py in __init__(self, costs, grad_vars, extra_vars, dtype, casting, compute_grads, **kwargs)
    689 
    690         if compute_grads:
--> 691             grad = tt.grad(self._cost_joined, self._vars_joined)
    692             grad.name = "__grad"
    693             outputs = [self._cost_joined, grad]

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in grad(cost, wrt, consider_constant, disconnected_inputs, add_names, known_grads, return_disconnected, null_gradients)
    637             assert g.type.dtype in theano.tensor.float_dtypes
    638 
--> 639     rval = _populate_grad_dict(var_to_app_to_idx, grad_dict, wrt, cost_name)
    640 
    641     for i in range(len(rval)):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in _populate_grad_dict(var_to_app_to_idx, grad_dict, wrt, cost_name)
   1438         return grad_dict[var]
   1439 
-> 1440     rval = [access_grad_cache(elem) for elem in wrt]
   1441 
   1442     return rval

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in <listcomp>(.0)
   1438         return grad_dict[var]
   1439 
-> 1440     rval = [access_grad_cache(elem) for elem in wrt]
   1441 
   1442     return rval

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_grad_cache(var)
   1391                     for idx in node_to_idx[node]:
   1392 
-> 1393                         term = access_term_cache(node)[idx]
   1394 
   1395                         if not isinstance(term, Variable):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_term_cache(node)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in <listcomp>(.0)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_grad_cache(var)
   1391                     for idx in node_to_idx[node]:
   1392 
-> 1393                         term = access_term_cache(node)[idx]
   1394 
   1395                         if not isinstance(term, Variable):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_term_cache(node)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in <listcomp>(.0)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_grad_cache(var)
   1391                     for idx in node_to_idx[node]:
   1392 
-> 1393                         term = access_term_cache(node)[idx]
   1394 
   1395                         if not isinstance(term, Variable):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_term_cache(node)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in <listcomp>(.0)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_grad_cache(var)
   1391                     for idx in node_to_idx[node]:
   1392 
-> 1393                         term = access_term_cache(node)[idx]
   1394 
   1395                         if not isinstance(term, Variable):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_term_cache(node)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in <listcomp>(.0)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_grad_cache(var)
   1391                     for idx in node_to_idx[node]:
   1392 
-> 1393                         term = access_term_cache(node)[idx]
   1394 
   1395                         if not isinstance(term, Variable):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_term_cache(node)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in <listcomp>(.0)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_grad_cache(var)
   1391                     for idx in node_to_idx[node]:
   1392 
-> 1393                         term = access_term_cache(node)[idx]
   1394 
   1395                         if not isinstance(term, Variable):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_term_cache(node)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in <listcomp>(.0)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_grad_cache(var)
   1391                     for idx in node_to_idx[node]:
   1392 
-> 1393                         term = access_term_cache(node)[idx]
   1394 
   1395                         if not isinstance(term, Variable):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_term_cache(node)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in <listcomp>(.0)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_grad_cache(var)
   1391                     for idx in node_to_idx[node]:
   1392 
-> 1393                         term = access_term_cache(node)[idx]
   1394 
   1395                         if not isinstance(term, Variable):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_term_cache(node)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in <listcomp>(.0)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_grad_cache(var)
   1391                     for idx in node_to_idx[node]:
   1392 
-> 1393                         term = access_term_cache(node)[idx]
   1394 
   1395                         if not isinstance(term, Variable):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_term_cache(node)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in <listcomp>(.0)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_grad_cache(var)
   1391                     for idx in node_to_idx[node]:
   1392 
-> 1393                         term = access_term_cache(node)[idx]
   1394 
   1395                         if not isinstance(term, Variable):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_term_cache(node)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in <listcomp>(.0)
   1059             inputs = node.inputs
   1060 
-> 1061             output_grads = [access_grad_cache(var) for var in node.outputs]
   1062 
   1063             # list of bools indicating if each output is connected to the cost

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_grad_cache(var)
   1391                     for idx in node_to_idx[node]:
   1392 
-> 1393                         term = access_term_cache(node)[idx]
   1394 
   1395                         if not isinstance(term, Variable):

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\gradient.py in access_term_cache(node)
   1218                             )
   1219 
-> 1220                 input_grads = node.op.L_op(inputs, node.outputs, new_output_grads)
   1221 
   1222                 if input_grads is None:

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\tensor\elemwise.py in L_op(self, inputs, outs, ograds)
    562 
    563         # compute grad with respect to broadcasted input
--> 564         rval = self._bgrad(inputs, outs, ograds)
    565 
    566         # TODO: make sure that zeros are clearly identifiable

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\tensor\elemwise.py in _bgrad(self, inputs, outputs, ograds)
    666                 ret.append(None)
    667                 continue
--> 668             ret.append(transform(scalar_igrad))
    669 
    670         return ret

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\tensor\elemwise.py in transform(r)
    657                 return DimShuffle((), ["x"] * nd)(res)
    658 
--> 659             new_r = Elemwise(node.op, {})(*[transform(ipt) for ipt in node.inputs])
    660             return new_r
    661 

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\tensor\elemwise.py in <listcomp>(.0)
    657                 return DimShuffle((), ["x"] * nd)(res)
    658 
--> 659             new_r = Elemwise(node.op, {})(*[transform(ipt) for ipt in node.inputs])
    660             return new_r
    661 

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\tensor\elemwise.py in transform(r)
    657                 return DimShuffle((), ["x"] * nd)(res)
    658 
--> 659             new_r = Elemwise(node.op, {})(*[transform(ipt) for ipt in node.inputs])
    660             return new_r
    661 

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\graph\op.py in __call__(self, *inputs, **kwargs)
    251 
    252         if config.compute_test_value != "off":
--> 253             compute_test_value(node)
    254 
    255         if self.default_output is not None:

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\graph\op.py in compute_test_value(node)
    124 
    125     # Create a thunk that performs the computation
--> 126     thunk = node.op.make_thunk(node, storage_map, compute_map, no_recycling=[])
    127     thunk.inputs = [storage_map[v] for v in node.inputs]
    128     thunk.outputs = [storage_map[v] for v in node.outputs]

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\graph\op.py in make_thunk(self, node, storage_map, compute_map, no_recycling, impl)
    632             )
    633             try:
--> 634                 return self.make_c_thunk(node, storage_map, compute_map, no_recycling)
    635             except (NotImplementedError, MethodNotDefined):
    636                 # We requested the c code, so don't catch the error.

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\graph\op.py in make_c_thunk(self, node, storage_map, compute_map, no_recycling)
    599                 raise NotImplementedError("float16")
    600         outputs = cl.make_thunk(
--> 601             input_storage=node_input_storage, output_storage=node_output_storage
    602         )
    603         thunk, node_input_filters, node_output_filters = outputs

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\link\c\basic.py in make_thunk(self, input_storage, output_storage, storage_map)
   1202         init_tasks, tasks = self.get_init_tasks()
   1203         cthunk, module, in_storage, out_storage, error_storage = self.__compile__(
-> 1204             input_storage, output_storage, storage_map
   1205         )
   1206 

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\link\c\basic.py in __compile__(self, input_storage, output_storage, storage_map)
   1140             input_storage,
   1141             output_storage,
-> 1142             storage_map,
   1143         )
   1144         return (

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\link\c\basic.py in cthunk_factory(self, error_storage, in_storage, out_storage, storage_map)
   1632             for node in self.node_order:
   1633                 node.op.prepare_node(node, storage_map, None, "c")
-> 1634             module = get_module_cache().module_from_key(key=key, lnk=self)
   1635 
   1636         vars = self.inputs + self.outputs + self.orphans

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\link\c\cmodule.py in module_from_key(self, key, lnk)
   1189             try:
   1190                 location = dlimport_workdir(self.dirname)
-> 1191                 module = lnk.compile_cmodule(location)
   1192                 name = module.__file__
   1193                 assert name.startswith(location)

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\link\c\basic.py in compile_cmodule(self, location)
   1548                     lib_dirs=self.lib_dirs(),
   1549                     libs=libs,
-> 1550                     preargs=preargs,
   1551                 )
   1552             except Exception as e:

C:\ProgramData\Anaconda3\envs\pm3env\lib\site-packages\theano\link\c\cmodule.py in compile_str(module_name, src_code, location, include_dirs, lib_dirs, libs, preargs, py_module, hide_symbols)
   2545             compile_stderr = compile_stderr.replace("\n", ". ")
   2546             raise Exception(
-> 2547                 f"Compilation failed (return status={status}): {compile_stderr}"
   2548             )
   2549         elif config.cmodule__compilation_warning and compile_stderr:

Exception: ("Compilation failed (return status=1): C:\\Users\\sreedatta\\AppData\\Local\\Theano\\compiledir_Windows-10-10.0.19041-SP0-Intel64_Family_6_Model_140_Stepping_1_GenuineIntel-3.7.9-64\\tmp81eb_5sq\\mod.cpp: In member function 'int {anonymous}::__struct_compiled_op_m67599e776bb0a5edbe20464e4ef6902fada5652e9f038845aa3f408620203691::run()':. C:\\Users\\sreedatta\\AppData\\Local\\Theano\\compiledir_Windows-10-10.0.19041-SP0-Intel64_Family_6_Model_140_Stepping_1_GenuineIntel-3.7.9-64\\tmp81eb_5sq\\mod.cpp:506:39: warning: narrowing conversion of 'V5_n0' from 'npy_intp' {aka 'long long int'} to 'int' inside { } [-Wnarrowing].      int init_totals[2] = {V5_n0, V1_n1};.                                        ^. C:\\Users\\sreedatta\\AppData\\Local\\Theano\\compiledir_Windows-10-10.0.19041-SP0-Intel64_Family_6_Model_140_Stepping_1_GenuineIntel-3.7.9-64\\tmp81eb_5sq\\mod.cpp:506:39: warning: narrowing conversion of 'V1_n1' from 'npy_intp' {aka 'long long int'} to 'int' inside { } [-Wnarrowing]. C:\\Users\\sreedatta\\AppData\\Local\\Theano\\compiledir_Windows-10-10.0.19041-SP0-Intel64_Family_6_Model_140_Stepping_1_GenuineIntel-3.7.9-64\\tmp81eb_5sq\\mod.cpp:521:5: warning: narrowing conversion of 'V5_stride0' from 'ssize_t' {aka 'long long int'} to 'int' inside { } [-Wnarrowing].      };.      ^. C:\\Users\\sreedatta\\AppData\\Local\\Theano\\compiledir_Windows-10-10.0.19041-SP0-Intel64_Family_6_Model_140_Stepping_1_GenuineIntel-3.7.9-64\\tmp81eb_5sq\\mod.cpp:521:5: warning: narrowing conversion of 'V1_stride0' from 'ssize_t' {aka 'long long int'} to 'int' inside { } [-Wnarrowing]. C:\\Users\\sreedatta\\AppData\\Local\\Theano\\compiledir_Windows-10-10.0.19041-SP0-Intel64_Family_6_Model_140_Stepping_1_GenuineIntel-3.7.9-64\\tmp81eb_5sq\\mod.cpp:521:5: warning: narrowing conversion of 'V1_stride1' from 'ssize_t' {aka 'long long int'} to 'int' inside { } [-Wnarrowing]. At global scope:. cc1plus.exe: warning: unrecognized command line option '-Wno-c++11-narrowing'. C:\\Users\\SREEDATTA\\AppData\\Local\\Temp\\ccjNNew1.s: Assembler messages:\r. C:\\Users\\SREEDATTA\\AppData\\Local\\Temp\\ccjNNew1.s:4410: Error: invalid register for .seh_savexmm\r. ", 'FunctionGraph(Elemwise{mul}(<TensorType(float64, col)>, <TensorType(int8, (True, True))>))')

Can one of you help?

Sree

Hello,

first of all, thanks for the great work on this project, I've been using bambi a lot and it has been super helpful!

I'm currently facing a (potential) issue when trying to make out-of-sample predictions for a logistic regression model built with the following formula:

y ~ x1 + x2 + x3 + (0 + x2|x1) + (0 + x3|x1)

where x1 and x2 are categorical variables with two dimensions respectively and x3 is a continuous variable.

The out-of-sample data I'm trying to make predictions for looks like this (exemplary):

| x1 | x2 | x3 | |----|----|-----| | 0 | 0 | 0 | | 0 | 0 | 0.5 | | 0 | 0 | 1 | | 0 | 0 | 1.5 | | 1 | 0 | 0 | | 1 | 0 | 0.5 | | 1 | 0 | 1 | | 1 | 0 | 1.5 |

There was no error when running model.predict(iData, data=out_of_sample_data, kind='mean'), however the spaghetti plot I generated from the posterior predictions looked off for when x1==1, the variance was much bigger than I expected. (I noticed this because I had manually made a plot displaying the 0.5 decision boundary for x3, i.e. the mean value and hdi intervals of x3 where there is a 50% probability of a positive outcome and that didn't match what I saw in the spaghetti plot.)

I then had a look at the code and noticed that the Z matrix generated in the predict method in models.py looked different from what I expected. Here's the code bit I'm referring to (last line):

        if self._design.group:
            if in_sample:
                Z = self._design.group.design_matrix
            else:
                Z = self._design.group._evaluate_new_data(data).design_matrix

What I got for Z was the following:

||||||| |---|---|---|---|-----|-----| | 1 | 0 | 0 | 0 | 0 | 0 | | 1 | 0 | 0 | 0 | 0.5 | 0 | | 1 | 0 | 0 | 0 | 1 | 0 | | 1 | 0 | 0 | 0 | 1.5 | 0 | | 0 | 0 | 1 | 0 | 0 | 0 | | 0 | 0 | 1 | 0 | 0 | 0.5 | | 0 | 0 | 1 | 0 | 0 | 1 | | 0 | 0 | 1 | 0 | 0 | 1.5 |

...but what I was expecting (after trying to make sense of it) was this:

||||||| |---|---|---|---|-----|-----| | 1 | 0 | 0 | 0 | 0 | 0 | | 1 | 0 | 0 | 0 | 0.5 | 0 | | 1 | 0 | 0 | 0 | 1 | 0 | | 1 | 0 | 0 | 0 | 1.5 | 0 | | 0 | 1 | 0 | 0 | 0 | 0 | | 0 | 1 | 0 | 0 | 0 | 0.5 | | 0 | 1 | 0 | 0 | 0 | 1 | | 0 | 1 | 0 | 0 | 0 | 1.5 |

so basically the second and third column swapped. I then added the following line:

Z[:, [1, 2]] = Z[:, [2, 1]]

to achieve that and the spaghetti plot I then generated matched my expectation.

It would be great if someone had a look at this and fixed it properly (if it really is an issue and not me making a mistake), I hope it was clear enough and if not, let me know!

Trying to mamba install bambi into an existing environment:

  + astor                 0.8.1  pyh9f0ad1d_0         conda-forge/noarch        Cached
  + aws-c-auth           0.6.21  hd46a2a8_1           conda-forge/osx-arm64     Cached
  + aws-c-compression    0.2.16  ha56e2a8_0           conda-forge/osx-arm64     Cached
  + aws-c-http           0.6.29  hb329ca4_0           conda-forge/osx-arm64     Cached
  + aws-c-mqtt           0.7.13  h5519404_10          conda-forge/osx-arm64     Cached
  + aws-c-s3              0.2.1  hb7b86b9_2           conda-forge/osx-arm64     Cached
  + aws-c-sdkutils        0.1.7  ha56e2a8_0           conda-forge/osx-arm64     Cached
  + aws-crt-cpp         0.18.16  h8bc873b_5           conda-forge/osx-arm64     Cached
  + bambi                 0.9.3  pyhd8ed1ab_0         conda-forge/noarch        Cached
  + base58                2.1.1  pyhd8ed1ab_0         conda-forge/noarch        Cached
  + boto3               1.26.37  pyhd8ed1ab_0         conda-forge/noarch        Cached
  + botocore            1.29.37  pyhd8ed1ab_0         conda-forge/noarch        Cached
  + formulae              0.3.4  pyhd8ed1ab_0         conda-forge/noarch        Cached
  + jmespath              1.0.1  pyhd8ed1ab_0         conda-forge/noarch        Cached
  + libarrow             10.0.1  hc5f4219_3_cpu       conda-forge/osx-arm64     Cached
  + libgrpc              1.51.1  h55edf5b_0           conda-forge/osx-arm64     Cached
  + s3transfer            0.6.0  pyhd8ed1ab_0         conda-forge/noarch        Cached

  Change:
────────────────────────────────────────────────────────────────────────────────────────

  - cryptography         38.0.4  py310h4fe9c50_0      conda-forge
  + cryptography         38.0.4  py310hfc83b78_0      conda-forge/osx-arm64     Cached
  - curl                 7.87.0  hbe9bab4_0           conda-forge
  + curl                 7.87.0  h9049daf_0           conda-forge/osx-arm64     Cached
  - hdf5                 1.12.2  nompi_h55deafc_101   conda-forge
  + hdf5                 1.12.2  nompi_ha7af310_101   conda-forge/osx-arm64     Cached
  - krb5                 1.20.1  h127bd45_0           conda-forge
  + krb5                 1.20.1  h69eda48_0           conda-forge/osx-arm64     Cached
  - libcurl              7.87.0  hbe9bab4_0           conda-forge
  + libcurl              7.87.0  h9049daf_0           conda-forge/osx-arm64     Cached
  - libevent             2.1.10  hbae9a57_4           conda-forge
  + libevent             2.1.10  h7673551_4           conda-forge/osx-arm64     Cached
  - libnghttp2           1.47.0  h232270b_1           conda-forge
  + libnghttp2           1.47.0  h519802c_1           conda-forge/osx-arm64     Cached
  - libssh2              1.10.0  hb80f160_3           conda-forge
  + libssh2              1.10.0  h7a5bd25_3           conda-forge/osx-arm64     Cached
  - libthrift            0.16.0  h1a74c4f_2           conda-forge
  + libthrift            0.16.0  h6635e49_2           conda-forge/osx-arm64     Cached
  - libzip                1.9.2  h96606af_1           conda-forge
  + libzip                1.9.2  h76ab92c_1           conda-forge/osx-arm64     Cached
  - python               3.10.8  hf452327_0_cpython   conda-forge
  + python               3.10.8  h3ba56d0_0_cpython   conda-forge/osx-arm64     Cached
  - sigtool               0.1.3  h7747421_0           conda-forge
  + sigtool               0.1.3  h44b9a77_0           conda-forge/osx-arm64     Cached

  Upgrade:
────────────────────────────────────────────────────────────────────────────────────────

  - arrow-cpp             9.0.0  py310h5547a8d_2_cpu  conda-forge
  + arrow-cpp            10.0.1  h44b9a77_3_cpu       conda-forge/osx-arm64     Cached
  - aws-c-cal            0.5.11  h4530763_0           conda-forge
  + aws-c-cal            0.5.20  h7a1267a_3           conda-forge/osx-arm64     Cached
  - aws-c-common          0.6.2  h3422bc3_0           conda-forge
  + aws-c-common          0.8.5  h1a8c8d9_0           conda-forge/osx-arm64     Cached
  - aws-c-event-stream    0.2.7  h9972306_13          conda-forge
  + aws-c-event-stream   0.2.16  h828d2a8_0           conda-forge/osx-arm64     Cached
  - aws-c-io             0.10.5  hea86ef8_0           conda-forge
  + aws-c-io            0.13.11  h2ec9475_2           conda-forge/osx-arm64     Cached
  - aws-checksums        0.1.11  h487e1a8_7           conda-forge
  + aws-checksums        0.1.14  ha56e2a8_0           conda-forge/osx-arm64     Cached
  - aws-sdk-cpp         1.8.186  h392f50b_4           conda-forge
  + aws-sdk-cpp         1.9.379  he22300a_6           conda-forge/osx-arm64     Cached
  - grpc-cpp             1.46.4  haeec53e_7           conda-forge
  + grpc-cpp             1.51.1  h44b9a77_0           conda-forge/osx-arm64       21kB
  - libgoogle-cloud       2.1.0  hec15cb4_1           conda-forge
  + libgoogle-cloud       2.5.0  hcf11473_1           conda-forge/osx-arm64     Cached
  - libprotobuf          3.20.2  hb5ab8b9_0           conda-forge
  + libprotobuf         3.21.12  hb5ab8b9_0           conda-forge/osx-arm64     Cached
  - openssl              1.1.1s  h03a7124_1           conda-forge
  + openssl               3.0.7  h03a7124_1           conda-forge/osx-arm64     Cached
  - orc                   1.7.6  hb9d09c9_0           conda-forge
  + orc                   1.8.1  hef0d403_0           conda-forge/osx-arm64     Cached
  - protobuf             3.20.2  py310h0f1eb42_1      conda-forge
  + protobuf            4.21.12  py310h0f1eb42_0      conda-forge/osx-arm64      287kB
  - pyarrow               9.0.0  py310had0e577_2_cpu  conda-forge
  + pyarrow              10.0.1  py310ha7868e4_3_cpu  conda-forge/osx-arm64        3MB

  Downgrade:
────────────────────────────────────────────────────────────────────────────────────────

  - click                 8.1.3  unix_pyhd8ed1ab_2    conda-forge
  + click                 8.0.4  py310hbe9552e_0      conda-forge/osx-arm64      153kB
  - streamlit            1.16.0  pyhd8ed1ab_0         conda-forge
  + streamlit             1.9.0  pyhd8ed1ab_0         conda-forge/noarch        Cached

  Summary:

  Install: 17 packages
  Change: 12 packages
  Upgrade: 14 packages
  Downgrade: 2 packages

  Total download: 3MB

Some of the downgrades (like streamlit) should not happen, I also don't see why all these aws packages should be installed.

Hi All, Merry Christmas.

Bounds are working prior = {'x1':bmb.Prior("TruncatedNormal", mu=10,sigma=5,lower=0, upper=50)} bmb.model('y ~ x1 + x2 + x3')

Bounds are not working prior = {'x1|level':bmb.Prior('Normal',mu=bmb.Prior("TruncatedNormal", mu=10,sigma=5,lower=0, upper=50),sigma = 5)} bmb.model('y ~ x1|level + x2 + x3')

I am a noob. Any guidance is appreciated. Thanks

I don't know if anyone is using bambi has conduct meta-analyses, but if so I would love to see an example. As a potential example source, there's a very nice example using brms here. If not, I will try (and most likely fail) to recreate this example and then ask questions.

Seems the code used to generate the scatterplot is not working anymore

I'm trying this but I don't like the end result

idata.posterior.plot.scatter(
    x="1|Subject", y="Days|Subject",
    hue="Subject__factor_dim",
    hue_style="discrete",
    add_colorbar=False,
    add_legend=False,
    ax=ax
)

I have ArviZ 0.14.0 and xarray 2022.11.0

I've been chatting with @aloctavodia about adding a Governance structure and document to the project. The current governance of ArviZ looks appealing as a starting point.

Anyone who reads this and is interested in participating in the governance of Bambi, please comment here or send me a message. The ultimate goal is to make this a more transparent and robust project that doesn't rely completely, and by (lack of) design, on one or two self-selected core developers.