What’s New¶
Historically pymer4
versioning was a bit all over the place but has settled down since 0.5.0. This page includes the most notable updates between versions but github is the best place to checkout more details and releases.
0.8.1¶
- Compatibility Updates:
This version includes a
noarch
build that should be installable on arm-based macOS platforms (e.g. M1, M2, etc)This version drops support for Python 3.7 and adds support for 3.9-3.11
- Breaking changes:
This version also uses
joblib
for model saving and loading and drops supported hdf5 files previously handled with thedeepdish
library as it is no longer actively maintained. This means that 0.8.1 will not be able to load models saved with earlier versions ofpymer4
!
0.8.0¶
- NOTE:
there was no 0.7.9 release as there were enough major changes to warrant a new minor release version
this version unpins the maximum versions of
rpy2
andpandas
if there are install issues with the
conda
release accompanying this version you should be able to successfully install into a conda environment using pip with the following:conda install 'r-lmerTest' 'r-emmeans' rpy2 -c conda-forge
followed bypip install pymer4
- New features:
Lm
models now supportfamily='binomial'
and uses theLogisticRegression
class from scikit-learn with no regularization for estimation. Estimates and errors have been verified against theglm
implementation in Rnew
lrt
function for estimating likelihood-ratio tests betweenLmer
models thanks to @dramanica. This replicates the functionality ofanova()
in R forlmer
models.new
.confint()
method forLmer
models thanks to @dramanica. This allows computing confidence intervals on 1 or more paramters of an already fit model including random effects which are not computed by default when calling.fit()
0.7.8¶
Maintenance release that pins
rpy2 >= 3.4.5,< 3.5.1
due to R -> Python dataframe conversion issue on recentrpy2
versions that causes a recursion error.Pending code changes to support
rpy2 >= 3.5.1
are tracked on this development branch. Upcoming releases will drop support forrpy2 < 3.5.X
Clearer error message when making circular predictions using
Lmer
models
0.7.7¶
This version is identical to 0.7.6 but supports
R >= 4.1
Installation is also more flexible and includes instructions for using
conda-forge
and optimized libraries (MKL) for Intel CPUs
0.7.6¶
- Bug fixes:
fixes an issue in which a
Lmer
model fit using categorical predictors would be unable to use.predict
or would return fitted values instead of predictions on new data. random effect and fixed effect index names were lost thanks to Mario Leaonardo Salinas for discovering this issue
0.7.5¶
This version is identical to 0.7.4 and simply exists because a naming conflict that resulted in a failed released to Anaconda cloud. See release notes for 0.7.4 below
0.7.4¶
- Compatibility updates:
This version drops official support for Python 3.6 and adds support for Python 3.9. While 3.6 should still work for the most part, development support and testing against this version of Python will no longer continue moving forward.
- New features:
utils.result_to_table
function nicely formats themodel.coefs
output for a fitted model. The docstring also contains instructions on using this in conjunction with the gspread-pandas library for “exporting” model results to a google sheet
0.7.3¶
- Bug fixes:
fix issue in which random effect and fixed effect index names were lost thanks to @jcheong0428 and @Shotgunosine for the quick PRs!
0.7.2¶
- Bug fixes:
fix bug in which
boot_func
would fail iwthy=None
andpaired=False
- Compatibility updates:
add support for
rpy2>=3.4.3
which handles model matrices differentlypin maximum
pandas<1.2
. This is neccesary until our other dependencydeepdish
adds support. See this issue
0.7.1¶
- Pymer4 will be on conda as of this release!
install with
conda install -c ejolly -c defaults -c conda-forge pymer4
This should make installation much easier
Big thanks to Tom Urbach for assisting with this!
- Bug fixes:
design matrix now handles rfx only models properly
compatibility with the latest version of pandas and rpy2 (as of 08/20)
Lmer.residuals
now save as numpy array rather thanR FloatVector
- New features:
stats.tost_equivalence
now takes aseed
argument for reproducibility
- Result Altering Change:
Custom contrasts in
Lmer
models are now expected to be specified in human readable format. This should be more intuitive for most users and is often what users expect from R itself, even though that’s not what it actually does! R expects custom contrasts passed to thecontrasts()
function to be the inverse of the desired contrasts. See this vignette for more info.In
Pymer4
, specifying the following contrasts:model.fit(factors = {"Col1": {'A': 1, 'B': -.5, 'C': -.5}}))
will estimate the difference between A and the mean of B and C as one would expect. Behind the scenes,Pymer4
is performing the inversion operation automatically for R.
Lots of other devops changes to make testing, bug-fixing, development, future releases and overall maintenance much easier. Much of this work has been off-loaded to automated testing and deployment via Travis CI.
0.7.0¶
dropped support for versions of
rpy2 < 3.0
Result Altering Change:
Lm
standard errors are now computed using the square-root of the adjusted mean-squared-error(np.sqrt(res.T.dot(res) / (X.shape[0] - X.shape[1])))
rather than the standard deviation of the residuals with DOF adjustment(np.std(res, axis=0, ddof=X.shape[1]))
. While these produce the same results if an intercept is included in the model, they differ slightly when an intercept is not included. Formerly in the no-intercept case, results from pymer4 would differ slightly from R or statsmodels. This change ensures the results are always identical in all cases.Result Altering Change:
Lm
rsquared and adjusted rsquared now take into account whether an intercept is included in the model estimation and adjust accordingly. This is consistent with the behavior of R and statsmodelsResult Altering Change: hc1 is the new default robust estimator for
Lm
models, changed from hc0API change: all model residuals are now saved in the
model.residuals
attribute and were formerly saved in themodel.resid
attribute. This is to maintain consistency withmodel.data
column names.New feature: addition of
pymer4.stats
module for various parametric and non-parametric statistics functions (e.g. permutation testing and bootstrapping)New feature: addition of
pymer4.io
module for saving and loading models to diskNew feature: addition of
Lm2
models that can perform multi-level modeling by first estimating a separate regression for each group and then performing inference on those estimates. Can perform inference on first-level semi-partial and partial correlation coefficients instead of betas too.New feature: All model classes now have the ability to rank transform data prior to estimation, see the rank argument of their respective
.fit()
methods.- New features for Lm models:
Lm
models can transform coefficients to partial or semi-partial correlation coefficientsLm
models can also perform weight-least-squares (WLS) regression given the weights argument to.fit()
, with optional dof correction via Satterthwaite approximation. This is useful for categorical (e.g. group) comparison where one does not want to assume equal variance between groups (e.g. Welch’s t-test). This remains an experimental featureLm
models can compute hc1 and hc2 robust standard errors
New documentation look: the look and feel of the docs site has been completely changed which should make getting information much more accessible. Additionally, overview pages have now been turned into downloadable tutorial jupyter notebooks
All methods/functions capable of parallelization now have their default
n_jobs
set to 1 (i.e. no default parallelization)Various bug fixes to all models
Automated testing on travis now pins specific r and r-package versions
Switched from lsmeans to emmeans for post-hoc tests because lsmeans is deprecated
Updated interactions with rpy2 api for compatibility with version 3 and higher
Refactored package layout for easier maintainability
0.6.0¶
Dropped support for Python 2
upgraded
rpy2
dependency versionAdded conda installation instructions
Accepted JOSS version
0.5.0¶
Lmer
models now support all generalized linear model family types supported by lme4 (e.g. poisson, gamma, etc)Lmer
models now support ANOVA tables with support for auto-orthogonalizing factors using the.anova()
methodTest statistic inference for
Lmer
models can now be performed via non-parametric permutation tests that shuffle observations within clustersLmer.fit(factors={})
arguments now support custom arbitrary contrastsNew forest plots for visualizing model estimates and confidence intervals via the
Lmer.plot_summary()
methodMore comprehensive documentation with examples of new features
Submission to JOSS
0.4.0¶
Added
.post_hoc()
method toLmer
modelsAdded
.simulate()
method toLmer
modelsSeveral bug fixes for Python 3 compatibility
0.3.2¶
addition of
simulate
module
0.2.2¶
Official pyipi release
0.2.1¶
Support for standard linear regression models
Models include support for robust standard errors, boot-strapped CIs, and permuted inference
0.2.0¶
Support for categorical predictors, model predictions, and model plots
0.1.0¶
Linear and Logit multi-level models