-
The code is spread between `items/cross_validation_item.py` and `sklearn/cross_validation.py`.
I find it pretty hard to contribute and to maintain.
I recommend to centralize the logic in a `sklear…
-
### Is your feature request related to a problem? Please describe.
Current way to access cross_validate is a bit weird for users (skore 0.4.0). It does things on their behalf that they do not reall…
-
### What is the desired addition or change?
It is envisioned to add a third Cross Validation (CV) class that purges and embargos data from the training sets after the k-fold splits to avoid…
-
## UX changes
Suggestion to avoid asking the user to do "return_xxx", asking them to give X and y again, and to allow for adding scorings as the current feature does.
With this piece of code, su…
-
First of all, thank you for the fantastic work on the sktime library! I am using the StackingForecaster in combination with ForecastingGridSearchCV for hyperparameter tuning. However, I'm unsure which…
-
### Is your feature request related to a problem? Please describe.
I think that it's not 100% clear what t…
-
# Summary
As part of our last Retrospective we've been discussing the ability to use the same suite to test against Lotus, Forest and whatever else is going to implement the node RPC API.
This …
-
### Issue
In our internal evals, we have seen a substantial improvement from alternating models when an oracle fails. In other words:
1. call sonnet with prompt
2. sonnet produces code that fails…
-
The cross-validation procedure currently implemented for finding the optimal density contrast value is ineffective when there is any significant regional field. It always overestimates the density co…
-
Add time series cross validation. https://github.com/FBruzzesi/timebasedcv
- [x] TimeSeriesCV(): https://business-science.github.io/pytimetk/reference/TimeSeriesCV.html
- [x] TimeSeriesCVSplitter(…