-
Hi. I've been experimenting with the new `recursive()` function quite a lot for panel data and I think the calibration process might not be working properly when using a recipe specification, maybe I'…
Teett updated
3 years ago
-
Using the development version of `modeltime`, I'm getting a hangup when combining a `workflow` with `add_formula()` where the formula has a date feature. It doesn't seem to be getting passed to the un…
-
In fact, this feature would have to be implemented in **`modeltime`** package, but I decided to put it here.
_Recursive ensemble_ instead of _ensemble of recursive models_ may have two potential cons…
-
I'm developing a [tidymodels extension package, `modeltime`](https://github.com/business-science/modeltime). Package development was going well but I'm running into an error where `parsnip` prevents `…
-
Hi Max & company,
Great R package. A few of my students are having issues installing parsnip-adjacent packages like `modeltime` and `treesnip`. [Here's a stackoverflow.](https://stackoverflow.com/…
-
I am trying to set up `modeltime.gluonts` for hyperparameter tuning. I am running the following code snippet:
```R
library(modeltime.gluonts)
model_spec_tune %
set_engine("gluonts_deepar")
…
-
Filing this issue specific to the `default_forecast_accuracy_metric_set()` used in Modeltime. This is related to tidymodels/yardstick#206.
-
I have a user of a developmental package `modeltime.gluonts` that has an error, which they posted here (business-science/modeltime.gluonts#9). I can trace the issue back to `parsnip`. In `new_model_sp…
-
This issue happens with `glmnet` models only.
```
Error in h(simpleError(msg, call)) :
error in evaluating the argument 'x' in selecting a method for function 'as.matrix': invalid class 'NA' …
-
Hello @mdancho84 , Thanks for the nice work!. It would be very good if you add H2O packages such as AutoML H2O. That also allows for the ensemble. If you can combine it with your ensemble average, …