Skip to content

Pull requests: sktime/pytorch-forecasting

New pull request New
34 Open 1,011 Closed
34 Open 1,011 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

[ENH] add feature scaling support for EncoderDecoderDataModule
#1983 opened Oct 12, 2025 by PranavBhatP • Draft
2 of 3 tasks
1
[ENH] Added metadata Testing for models and data modules for ptf-v2
#1973 opened Sep 20, 2025 by ParamThakkar123 Loading…
3 tasks done
2
[ENH] Duet Implementation
#1968 opened Sep 8, 2025 by Shvrth • Draft
3 tasks done
1
13
[ENH] Refactor _loss_mapping
#1967 opened Sep 8, 2025 by ParamThakkar123 Loading…
3 tasks done
1
78
[Docs] Auto-generate model overview from registry.
#1958 opened Aug 28, 2025 by Pinaka07 Loading…
10
[ENH] Consistent 3D output for single-target point predictions in TimeXer v1. enhancement New feature or request
#1936 opened Aug 1, 2025 by PranavBhatP Loading…
2 of 3 tasks
@PranavBhatP
57
[MNT] deduplicating utilities with scikit-base enhancement New feature or request maintenance Continuous integration, unit testing & package distribution
#1929 opened Jul 25, 2025 by fkiraly Loading…
[ENH] Refactor metrics to be polymorphic enhancement New feature or request module:models
#1897 opened Jun 19, 2025 by phoeenniixx • Draft
@phoeenniixx
23
[ENH] [WIP] Standardize model output to 4d tensor enhancement New feature or request module:models
#1895 opened Jun 18, 2025 by PranavBhatP Loading…
2 of 3 tasks
1
@PranavBhatP
33
[ENH] Add warning for multi-target forecasting support in BaseModel
#1880 opened Jun 9, 2025 by PranavBhatP Loading…
2 of 3 tasks
1
6
[WIP] [ENH] Add timeseries integration test for various models enhancement New feature or request
#1875 opened Jun 5, 2025 by jobs-git Loading…
2
[ENH] Precompute data to accelerate training in GPU enhancement New feature or request
#1850 opened May 27, 2025 by jobs-git Loading…
8 tasks done
8
39
[ENH] Allow direct initialization of NBeats model without from_dataset
#1837 opened May 19, 2025 by MeherBhaskar Loading…
3 of 4 tasks
7
[DO NOT MERGE] Experimental PR to demonstrate TimeXer model usage with v2 design
#1830 opened May 12, 2025 by PranavBhatP • Draft
4 tasks
@PranavBhatP
13
[MNT] in CI check for missing init files maintenance Continuous integration, unit testing & package distribution
#1801 opened Mar 15, 2025 by fkiraly • Draft
[ENH] Informer Implementation
#1796 opened Mar 13, 2025 by Ankit-1204 Loading…
2 of 4 tasks
@Ankit-1204
2
[BUG] pytorch-forecasting#1752 Fixing bug Something isn't working
#1786 opened Feb 26, 2025 by RUPESH-KUMAR01 Loading…
1 of 4 tasks
5
[ENH] speculative design for raw data container DataSet v2 API design API design & software architecture enhancement New feature or request
#1757 opened Jan 22, 2025 by fkiraly • Draft
1
[ENH] speculative design for raw data container DataSet enhancement New feature or request
#1755 opened Jan 16, 2025 by fkiraly • Draft
2
Fix examples
#1589 opened Aug 4, 2024 by freesoul Loading…
4 tasks
[ENH] Support tuning any model and extend LSTMModel in docs to support multi-target datasets enhancement New feature or request
#1449 opened Nov 22, 2023 by svnv-svsv-jm Loading…
4
[BUG] Nhits weight argument fix in TimeSeriesDataSet
#1432 opened Nov 6, 2023 by manitadayon Loading…
5
[ENH] change image logging from matplotlib figure to binary image enhancement New feature or request
#1417 opened Oct 21, 2023 by broomva Loading…
4 tasks
1
TSMixer
#1375 opened Sep 10, 2023 by jdb78 Loading…
2 of 3 tasks
2
Bugfix: handle case where 'losses' is a list of tensors
#1357 opened Aug 4, 2023 by TeodorChiaburu Loading…
3
ProTip! Adding no:label will show everything without a label.