Research map / Time series / Foundation models for time series
PYRREGULAR: A Unified Framework for Irregular Time Series, with Classification Benchmarks
Francesco Spinnato, Cristiano Landi
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| store data in sparse COO format instead of raw | pyrregular · GS dataset | disk size (GB): 0.24 -> 0.09 | Improved |
| store data in sparse COO format instead of raw | pyrregular · TA dataset | disk size (GB): 1.81 -> 0.08 | Improved |
| use sparse COO instead of minimally ragged dense | pyrregular · PA2 dataset | memory (GB): 5.33 -> 3.93 | Improved |
| use sparse COO instead of fully ragged dense | pyrregular · TA dataset | memory (GB): 4135.02 -> 0.34 | Improved |
| fine-tune generalist ROCKET vs specialist MusicNet | ROCKET · P12 | AUC: 53.4 -> 86.1 | Hurt |
| use specialist deep model (MTSFormer) instead of LGBM | LGBM · P19 | AUC: 85.2 -> 88.8 | Improved |
| use LGBM instead of ContiFormer | ContiFormer · P19 | AUC: 79.2 -> 85.2 | Improved |
| use partially-observed-aware models (SAITS, BRITS) instead of ROCKET/BORF/LGBM | ROCKET · partially observed datasets | F1 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- ContiFormer — ContiFormer: Continuous-Time Transformer for Irregular Time Series Modeling (NeurIPS 2023) · better on 2 of 4 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Peri-midFormer: Periodic Pyramid Transformer for Time Series Analysis (NeurIPS 2024 · similarity 0.84)
- Time-IMM: A Dataset and Benchmark for Irregular Multimodal Multivariate Time Series (NeurIPS 2025 · similarity 0.83)
- TimeMixer++: A General Time Series Pattern Machine for Universal Predictive Analysis (ICLR 2025 · similarity 0.83)
- An Unsupervised Approach for Periodic Source Detection in Time Series (ICML 2024 · similarity 0.83)
- DualDynamics: Synergizing Implicit and Explicit Methods for Robust Irregular Time Series Analysis (AAAI 2025 · similarity 0.83)
- MOMENT: A Family of Open Time-series Foundation Models (ICML 2024 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
