We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
Add AutoRegressivePipeline class that applies model in autoregressive style.
According to Kaggle competitions there are numerous cases when applying regressor models in autoregressive style beats their classic kind.
Add class AutoRegressivePipeline. It should implement this interface:
AutoRegressivePipeline(model, transforms, horizon, step=1)
Where step show after how many steps features will be recalculated, also this value will be used as horizon parameter in TSDataset.make_future
step
No response
The text was updated successfully, but these errors were encountered:
AutoRegressivePipeline
Mr-Geekman
Successfully merging a pull request may close this issue.
🚀 Feature Request
Add AutoRegressivePipeline class that applies model in autoregressive style.
Motivation
According to Kaggle competitions there are numerous cases when applying regressor models in autoregressive style beats their classic kind.
Proposal
Add class AutoRegressivePipeline.
It should implement this interface:
Where
step
show after how many steps features will be recalculated, also this value will be used as horizon parameter in TSDataset.make_futureTest cases
No response
Alternatives
No response
Additional context
No response
Checklist
The text was updated successfully, but these errors were encountered: