-
The following error is raise in a `jupyter-lab` environment. `data_df` is unremarkable `pandas.DataFrame` whose first column is the grouping variable (as a string). Strangely this bug only raises some…
-
Two potential speed-ups for spline and HSGP models (without random effects)
## 1. use _glm more often.
E.g. formula `y ~ s(x)` generates code with
```
vector[N] mu = Intercept + rep_vecto…
-
Hi,
I am trying to do a multi dimensional input for a classification GP, it works very well for one dimension, but as soon I add another feature gives me a hashable error, I am pretty sure is somet…
-
-
Dear Enrico!
As requested, I will post my enhancement proposals here on github. Let's start with the generic example I gave. For example, I have 2 or more time points/groups and I want to predict s…
-
### I don't have 8 GPUS, so I chang3 Num_GPUS to 2 and it raise this error. How can I fix it?
I use e2e_cascade_rcnn_X-101-64x4d-FPN_1x.yaml. I change it like:
MODEL:
TYPE: generalized_rcnn
CO…
-
# LASSO regression using tidymodels and #TidyTuesday data for The Office | Julia Silge
I’ve been publishing screencasts demonstrating how to use the tidymodels framework, from first steps in modeling…
-
The families currently supported by projpred (`gaussian()`, `binomial()`, `brms::bernoulli()`, `poisson()`) belong to the class of "exponential dispersion (ED) families" (Jørgensen, 1987; see also sec…
-
On the FRES call last Friday, we spent a significant amount of time thinking through how we want to describe the initial conditions. The issues we have are:
* Whether what ASPECT sees are the initial…
-
#### Summary:
Match the RStan feature that lets you indicate which parameters to save by name (not by index within name).
I'm not sure if this is something that's best to filter down at the Stan…