WIP Documentation #358
WIP Documentation #358
Conversation
JulietteOrtholand
force-pushed
the
documentation
branch
from
c29e44b to
d423ede
Compare
| [200 rows x 4 columns] | ||
| ``` | ||
|
|
||
| __ more a bayesian one:__ random effects are estimated using a Gibbs sampler with an option on the burn-in phase [REF part?] and temperature scheme [REF part?]. Nevertheless, for now the package enable only to extract the mean or the mode of the posterior distribution. |
There was a problem hiding this comment.
@KaisaridiSofia Can you confirm you will explain temperature and burn-in in fit part so I can refer to this part here?
There was a problem hiding this comment.
Yes @JulietteOrtholand, I will do that
| __ more of a frequentist one:__ random effects are estimated using the solver [_minimise_](https://docs.scipy.org/doc/scipy/reference/generated/scipy.optimize.minimize.html) from the package Scipy (\cite{2020SciPy_NMeth}) to maximize the likelihood knowing the fixed effects. | ||
|
|
||
| ```python | ||
| >>> personalize_settings = AlgorithmSettings("scipy_minimize", seed=0) |
There was a problem hiding this comment.
It's a copy-paste from the nut-shell code, is it an issue? Should I make a reference to it instead of copy-pasting? I have copy-pasted because I found it clearer for the reader and more coherent with the following section. WDYT @caglayantuna and @NicolasGensollen ?
| __Make predictions:__ TADPOLE and others | ||
|
|
||
| ### References |
There was a problem hiding this comment.
Is there a clever way to handle citations @caglayantuna and @NicolasGensollen? Would you have some recommendations on that? I was planning to put hyperlinks but might not be the best for journal papers..
| @@ -6,4 +6,34 @@ | |||
|
|
|||
|
|
|||
| ## Riemanian framework | |||
There was a problem hiding this comment.
@KaisaridiSofia please let me know if you understand the Riemannian framework, if you do so, I'll ask the others if they do, as I guess you are more aware of it.
This PR is for the three documentation tasks I have to do: