-
binomial (logit/probit)
gaussian (identity)
negative binomial (logit/probit)
multinomial (logit/probit)
-
Some functions should be renamed with a `.` because they implement some internal behaviour (not for the user).
| Old function | New name | GitHub link |
|----|----|--------|
| `likelihood` | `.li…
-
I'm writing some code for a special case of generalized likelihood maximization. To do this, I have functions
score(noise::Distribution, theta, observations)
and
information(noise::Distribution, t…
-
These are the future steps Robyn and I agreed on September 4th 2024
**Immediate TODO**
- [ ] replace likelihood functions: Robyn
- [ ] built in testing functions -check in with Andy
- [ ] build …
-
We started implementing support the _WeightIt_ package in our easystats-eco-system, mainly in _insight_ (https://cran.r-project.org/package=insight) and _parameters_ (https://cran.r-project.org/packag…
-
This can be done in parallel to #4 , in order to try to find a good neural network architecture @melli1992
-
**Is your feature request related to a problem? Please describe.**
The user is limited to being able to define logprior and loglikelihood functions. This is limiting for many applications, although …
-
When using kriging, the most expensive task is the calibration of the parameters of the covariance function, done by log-likelihood maximization. When the optimization is done using a gradient-based a…
-
# Issue: Unify Functions with Similar Logic but Different Parameters
## Description
There are several functions in our codebase that perform the same or very similar operations but accept differ…
-
# 🚀 Feature Request
I would like to implement a changepoint kernel in my GP model. An example is given in [Duvenaud's thesis](https://www.cs.toronto.edu/~duvenaud/thesis.pdf) in Eq. 2.19
![Screens…