-
Study the impact of single and mixed precision on the performance predictions of the ML models. Need to understand from the data point of view used for training if single precision suffices ? or do we…
-
Would it be possible to include some more information in the returned objects from `svylme()`, or adding some more methods? Like `nobs()`, `family()`, maybe the call, the model frame, AIC, ... etc.? W…
-
We have made the choice of full and modified Newton method optional.
Full Newton method is still the default since FV is also still the default, for which the modified Newton showed mixed results wit…
-
**What needs to be cleaned up?**
Right now our data models are mixed up with our domain models. This eliminates duplicate code, but makes the models very brittle as any change to the domain model n…
-
It would be nice to have an analog to `coef()` for mixed effects models. This returns `fixef() + ranef()` for each random parameter. Modeling packages vary in whether SEs and intervals are provided fo…
-
```
For both cpu, and gpu implementations, including implementing derivatives.
```
Original issue reported on code.google.com by `daniel.a...@gmail.com` on 7 Jan 2010 at 2:44
-
### JASP Version
0.19 beta
### Bug Description
I scraped some modules for strings that are not yet translatable. Found them in:
- [ ] generalized linear mixed models
- [ ] linear mixed mo…
-
Hi,
I was wondering what precision these models were trained at? Float32? Float16? Bfloat16? With mixed precision?
Thanks.
-
Actually the model present in the [documentation](https://csia-pme.github.io/csia-pme/reference/engine/#uml-diagram), represent how we would like the inheritance and links between classes to be. But i…
-
Let's say that some samples are grouped (patients in a hospital, etc.) and that these groupings induce correlations between the samples.
If I'm interested in understanding the effect of treatment …