-
In Azure, we should be able to set a node as spot (preemptible) by setting [properties.priority = 'Spot'](https://docs.microsoft.com/en-us/rest/api/compute/virtualmachines/createorupdate#virtualmachin…
-
Sometimes it's useful to model variance of the distribution instead of standard deviation, because unlike standard deviations, a sum of two variances equals to the variance of a sum. An easy way to im…
-
I was hoping to find some terms for describing dependence found in longitudinal, repeated measures or time series data. The reason is that over in [NIDM](http://github.com/nidash/nidm) we've found th…
-
If the user adds a path to a base model, rather than just "does not recognize paths" we should tell them:
> for path models, you need to add `type= "RAM"`
Example and error below:
```R
dat…
-
Paul, I have been trying the cumulative() family with a 2PL item response model having 4 possible levels. However, I keep seeing the error that non-linear models are not allowed for ordinal models.
…
-
The logpdf function in the build_loss_objective is not set up properly.
What should take place? For a simple demo model like Lotka-Volterra or Lorenz and ONE DATASET, fit_mle should fit a MvNormal…
-
Hi Paul,
thanks again, for truly maximizing our choice of models so far.
Logitnormal distribution is an alternative to beta distribution and applies for response types with lower and upper boun…
-
With the implementation of distributional regression in the upcoming release of **brms** 1.0, I think it makes sense to think of additional response distributions (i.e., 'families') that could be wort…
-
The following minimal example fails:
```
brm(bf(prop ~ 1, phi ~ 1),
data = data.frame(prop = rbeta(100, shape1 = 2, shape2 = 2)),
family = Beta())
```
```
SYNTAX ERROR, MESSAGE(S) …
-
#13163 will be bringing in the long-awaited replacement of numpy's PRNG infrastructure. In the interest of keeping that PR manageable, we will merge it to master before all of the decisions are finali…
rkern updated
5 years ago