madtoinou

Issue with the validation_data with unsupervised and semi-supervised learning using generators

2

**Task** I am trying to obtain the validation loss of my model during the training (call to fit). We work with worms brain data, some of the volume are segmented....

fix: datetime_attribute account for 0 or 1-indexing of the attributes

1

Fixes #2219. ### Summary - `datetime_attribute_timeseries()` was not taking into account the indexing of the attribute (1 for month, day or week for example), resulting in a shift of the...

Fix: estimator getter and lagged_label_name

1

Fixes #2217. ### Summary - Fixed bug in `get_multioutput_estimator()` where the incorrect estimator could be returned - Added an attribute to store the lagged label, each one corresponding to an...

Checklist before merging this PR: - [ ] Mentioned all issues that this PR fixes or addresses. - [ ] Summarized the updates of this PR under **Summary**. - [...

**Is your feature request related to a current problem? Please describe.** `statsforecast` improved their AutoARIMA implementation, it's not possible to fit it on one series and apply it to another...

Fix/update doc

2

Checklist before merging this PR: - [X] Mentioned all issues that this PR fixes or addresses. - [X] Summarized the updates of this PR under **Summary**. - [x] Added an...

Feat/scalar with window

2

Checklist before merging this PR: - [x] Mentioned all issues that this PR fixes or addresses. - [x] Summarized the updates of this PR under **Summary**. - [x] Added an...