-
It would be useful to consider support for non-scalar `poutine.scale` with `TraceEnum_ELBO`. In general, importance sampling (such as stratified sampling) across different plates requires scaling diff…
-
You can train values so that an `m`-value ends up with a value of `NaN`. This then breaks the SQL that is generated in `.predict()`.
Here is a non-elegant reprex, very similar to #2333 (although no…
-
**Describe the bug**
When the levels of a categorical moderator are not capitalised to begin with, orchard_plot() breaks with changes to 'upper' or 'tree.order' options. Below I demonstrate with a re…
-
One-class classification models, such as SVM can be used for imbalanced data
Also consider data augmentation - generate similar habitable lanes for training the classifier and Down-sampling of non-ha…
-
I ran the trainer and got this error:
My system has 8GB VRAM and 32GB RAM. I have 5 pictures in the dataset. I used the default settings in gradio.
[2024-09-08 16:11:06] [INFO] subprocess.CalledPr…
-
This issue started as part of investigation of reasons for 100% CPU usage on Mikrotik boxes during DDoS. Particular deployment had no firewall rules and expected to handle all the traffic using Fast P…
-
#### Summary:
In a simple linear Gaussian example with p= and n=, the scale of data can affect the warmup time from around 1 second to more than 100 seconds. This is likely to be related to initial v…
-
tensor at non-singleton dimension 2
-
The algorithm for calculation of joint inclusion probabilities are different for different probabilistic sampling schemes that are not differentiated by the current codes for selectionMethod. Pairwise…
-
1. On what basis do comparison groups get constructed?
a. What criteria do we use to define the composition of a comparison group?
b. How are existing comparison group members disqualified and remov…