-
-
Notifications
You must be signed in to change notification settings - Fork 2.1k
[ENH] neural network libraries in thuml time-series-library #7243
Copy link
Copy link
Open
Labels
enhancementAdding new functionalityAdding new functionalityimplementing algorithmsImplementing algorithms, estimators, objects native to sktimeImplementing algorithms, estimators, objects native to sktimeinterfacing algorithmsInterfacing existing algorithms/estimators from third party packagesInterfacing existing algorithms/estimators from third party packagesmodule:forecastingforecasting module: forecasting, incl probabilistic and hierarchical forecastingforecasting module: forecasting, incl probabilistic and hierarchical forecasting
Description
Metadata
Metadata
Assignees
Labels
enhancementAdding new functionalityAdding new functionalityimplementing algorithmsImplementing algorithms, estimators, objects native to sktimeImplementing algorithms, estimators, objects native to sktimeinterfacing algorithmsInterfacing existing algorithms/estimators from third party packagesInterfacing existing algorithms/estimators from third party packagesmodule:forecastingforecasting module: forecasting, incl probabilistic and hierarchical forecastingforecasting module: forecasting, incl probabilistic and hierarchical forecasting
I found this repository
https://github.com/thuml/Time-Series-Library
which implements a large number of state-of-art neural network models in
torch, and seems to be updated regularly.The repository is not a pypi package and relies on source install.
There is also a large PR that turns it into a package, but it seems stale or unreviewed:
thuml/Time-Series-Library#457
I wonder about the best way to integrate here. Nothing keeps us from copying layers and networks, but that is perhaps not the best way to proceed (and also not that nice towards the authors). But interfacing is not possible without proper package management, similar to the
cure-labproblem.From a tech perspective, a merge or some kind of connection with
pytorch-forecastingcould make sense, and then interfacing fromsktimeone layer higher. A connection which is not merge is turningpytorch-forecastinginto atorchneural network and components indexer, usingscikit-base, where components are not necessarily directly in the package.FYI @XinyuWuu, the most active devs seem to be all from China, do you know them? I have also not reviewed the code in detail, e.g., intersections with
pytorch-forecasting, or withneuralforecast.