Changelog
0.7.0
New features
- Add
GroupSpecificTerm.eval_new_data_group_index() to distinguish fitted, missing, and new grouping levels when evaluating new data.
0.6.2
Maintenance and fixes
- Update versions (#122)
- Extend equality methods (#123)
0.6.1
Maintenance and fixes
- Remove
sparse dependency (#121)
Documentation
- Update README to mention formulae supports Python>=3.8 (#121)
0.6.0
New features
- Add
as_dataframe() method to GroupEffectsMatrix (#115)
- Use sparse matrices for group-specific design matrices (#117, #118, #119)
Maintenance and fixes
- Fix pandas version check (#110)
- Test on Python 3.10, 3.11, and 3.12 (#113)
- Migrate development tooling and workflow to pixi (#116)
Deprecation
- Drop support for Python 3.7; Python ≥ 3.8 is now required (#110)
0.5.4
Maintenance and fixes
- Make sure model_description always returns a Model instance (#112)
0.5.3
Maintenance and fixes
- Make formulae reach transformations before outer names (#109)
0.5.2
Maintenance and fixes
- Update is_categorical_dtype for pandas >= 2.10 (#105)
- Add automatic versioning to the library (#106)
- Interpret True, False, and None as Python literals (#107)
0.5.1
Maintenance and fixes
- Fix bug when intercept is inserted after categorical variable (#102)
0.5.0
New features
- The library now supports configuration variables (#95)
- Allow evaluations on new data with new categories (#96)
- Return ready to be used group-specific effects design matrix when there are new groups (#100)
0.4.0
New features
- Implement a @register_stateful_transform decorator to register new stateful transformations (#85)
Maintenance and fixes
- Fix evaluation of new offset terms with call arguments (#81)
- Multiple calls with unnamed attributes resulted in errors due to trying to hash unhashable types (#88)
- Term names with call components now use the proper lexeme when one argument is a string (#89)
- Detect stateful transformations by attribute instead of name (#90)
- Moved tests out of the
formulae directory (#91)
- Use pyproject and improve deployment workflow (#92)
v0.3.4
Maintenance and fixes
- Fixed a bug in the levels of interaction terms involving numeric terms with multiple columns (b4a1f73)
v0.3.3
Maintenance and fixes
- Fixed a bug in
CategoricalBox. Now it considers the order of the categories if data is ordered and levels is None (#73)
v0.3.2
Maintenance and fixes
- Fixed a bug in
CategoricalBox because it failed to convert categorical series to numpy arrays. Now it works. (#72)
v0.3.1
Maintenance and fixes
- Renamed
ResponseVector to ResponseMatrix (#71)
- Renamed
design_vector to design_matrix in ResponseMatrix(#71)
- Updated docstrings in
formulae/matrices.py (#71)
Deprecation
- Removed
binary and success attributes from ResponseMatrix (#71)
v0.3.0
New features
- We can create our own encodings such as Treatment or Sum encoding. These are subclasses of
Encoding.
- Added two aliases
T and S that are shorthands of C(var, Treatment) and C(var, Sum) respectively.
- DesignVector, CommonEffectsMatrix and GroupEffectsMatrix now retrieve their values when passed to
np.array() and np.asarray().
- Add
poly() stateful transform.
na_action in design_matrices() can be "pass" which means not to drop or raise error about missing values. It just keeps them (#69)design_matrices() gains a new argument extra_namespace. It allows us to pass a dictionary of transformations that are made available in the environment where the formula is evaluated (#70)
Maintenance and fixes
- Fixed a bug in the addition of lower order terms when the original higher order term wasn’t full-rank.
- Columns for categorical terms are integer by default. They are casted to float only if there are other float-valued columns in the design matrix.
- Updated str and repr methods of
ResponseVector, CommonEffectsMatrix, and GroupEffectsMatrix.
- Added str and repr methods for
DesignMatrices.
- Added get_item method for
DesignMatrices.
- Added support for comparison operators within function calls.
v0.2.0
- Added
bs(), a stateful transformation for basis splines (#52)
- Subset notation in response terms can now be an identifier. It is interpreted as a string (#52)
True, False and None are correctly evaluated in function calls arguments now (#53)- The
.set_type() on each term used to be called twice. Now we call it once (#53)
- Added the function
get_function_from_module(). Now we don’t use Python’s eval() anymore (#53)
v0.1.4
- Revert changes back to v0.1.2
v0.1.3
- Revert changes back to v0.1.1
v0.1.2
- Added
prop() function to handle response terms that a proportion computed from two variables (#40)
- Categorical responses can contain more than one level (#40)
- Added
binary() function to convert numeric/categorical variables into 0-1 variables (#41)
- Modified
C() to make it simpler to obtain categorical variables and manipulate their levels (#41)
- Added
offset() (#42)
v0.1.1
Maintenance and fixes
- Fix group-specific effects matrix shape (#35)
Documentation
- Add a Getting started section to webpage and remove that content from README.md (#36)
v0.1.0
New features
- Factor of group specific effect can be an interaction now (#31)
Maintenance and fixes
- Full interaction does not result in shared components anymore (#30)
- Add information about levels to interaction terms (#33)
Documentation
- Added Internals section to documentation (#28)
v0.0.10
Maintenance and fixes
- Nested stateful transofrmations now work as expected (#19)
Documentation
- Added a changelog file to track changes (#20)
- Formulae now has a github pages website (#21)