[ENH] Implement s2s Linear Forecaster models family#3171
Conversation
9567e61 to
2a74356
Compare
hadifawaz1999
left a comment
There was a problem hiding this comment.
I took a first look, a small comment to begin,i will check the rest
hadifawaz1999
left a comment
There was a problem hiding this comment.
LGTM ! very nice, @MatthewMiddlehurst @TonyBagnall @alexbanwell1 let me know if you have any comments before merging
|
Looks good to me. Can't comment on the actual model but the forecasting module changes and series to series mixin look sensible |
|
Also not familiar with the algorithm or whether this is a correct implementation, but would be good to discuss how to integrate series to series before putting this in probably. Thanks very much for the work though, this is helpful. |
|
I had a brief discussion on s2s and this linear forecaster with @hadifawaz1999 on discord. let me know whatever is decided and I refactor this accordingly. |
Reference Issues/PRs
Implements LinearForecaster family of models (linear, nlinear and dlinear) following the PR :
#3146
What does this implement/fix? Explain your changes.
This PR implements a simple 1 layered MLP based Linear Forecaster as a baseline for series2series deep learning forecasters.
Does your contribution introduce a new dependency? If yes, which one?
No
Any other comments?
PR checklist
For all contributions
For new estimators and functions
__maintainer__at the top of relevant files and want to be contacted regarding its maintenance. Unmaintained files may be removed. This is for the full file, and you should not add yourself if you are just making minor changes or do not want to help maintain its contents.For developers with write access