- patch release to fix numpyro integration issues. The default installation workflow now uses
numpyro==0.16.1
to avoidTypeError
when using a discrete data distribution, e.g. Bernoulli.
- patch release to fix installation issues. The default installation workflow now uses
scipy==0.12
to avoid errorcannot import name 'gaussian' from 'scipy.signal'
.
- patch release to fix installation issues. The default installation workflow now uses
numpyro==0.13.2
to avoidModuleNotFoundError: No module named 'jax.linear_util'
.
- Fixed bug where beta distribution was being treated as Laplace distribution.
- Fixed bug where nested plated were being indexed incorrectly in Python.
- Switched inference to Python's
numpyro
dag_greta()
is now deprecated;dag_numpyro()
should be used as a drop-in replacement.- Added helper function for Python dependencies:
install_causact_deps
. - Updated documentation to reflect switch to
numpyro
. - Added ability to choose fill of nodes during
dag_render()
. - Added
causact
probability distribution functions. See?causact::distributions
. - Temporarily removed support for the
dim
argument of probability distributions.
- Added `vignette("narrative-to-insight-with-causact") to introduce the package's encouraged user workflow.
- Allow
greta::variable()
to be used for flat priors.
- Fixed bugs introduced by changes in
tidyr::replace_na()
. - Fixed bug related to rendering generative DAGs where two variables were both on multiple plates.
- Support for using
greta
matrix operations like%*%
and othergreta
functions within therhs
argument of thedag_node
function has been added. - The readme has been updated to include a
mixture
example.
- Small graphs without an
rhs
argument supplied todag_node()
displayed adplyr
warning message. - Exported
meaningfulLabels()
function for users who are using code generated bydag_greta(mcmc=FALSE)
.
dag_render()
displayed weird DAGs when users havedplyr (>=v1.0)
installed. Isolated problem to changes indplyr::group_indices()
anddplyr::bind_cols
that caused a jumbling of node id's.- Updated README.md to display images and give CRAN install instructions.
- Initial release for CRAN
- Biggest change from previous development versions is that
dag_greta()
now returns a dataframe as opposed to creating variables in your global environment.