| ---
|
| license: cc-by-4.0
|
| task_categories:
|
| - time-series-forecasting
|
| tags:
|
| - tsfile
|
| - timeseries
|
| - time-series
|
| - classification
|
| pretty_name: SelfRegulationSCP2 (TsFile)
|
| size_categories:
|
| - 10K<n<100K
|
| configs:
|
| - config_name: default
|
| data_files:
|
| - split: train
|
| path: SelfRegulationSCP2_train.tsfile
|
| - split: test
|
| path: SelfRegulationSCP2_test.tsfile
|
| ---
|
|
|
| # SelfRegulationSCP2 (TsFile)
|
|
|
| Apache TsFile version of the `SelfRegulationSCP2` UEA classification subset of
|
| [`thuml/Time-Series-Library`](https://huggingface.co/datasets/thuml/Time-Series-Library).
|
|
|
| ## Overview
|
|
|
| EEG slow cortical potentials (ALS patient); classify cursor up vs down.
|
|
|
| - **Train samples:** 200 • **Test samples:** 180.
|
| - **Dimensions (channels):** 7 • **Series length:** 1152.
|
| - **Classes:** 2.
|
|
|
| Each sample is an independent multivariate series. TRAIN and TEST are stored as
|
| two separate TsFiles (`SelfRegulationSCP2_train.tsfile` / `SelfRegulationSCP2_test.tsfile`).
|
|
|
| ## Schema (TsFile structure)
|
|
|
| - **sample_index** (TAG) — one device per sample. Query one sample with
|
| `WHERE sample_index=0`.
|
| - **Time** (INT64) — within-sample position (0..1151); a frame index, not a
|
| wall-clock timestamp (the sktime `.ts` source has none).
|
| - **dim_0 … dim_6** (FIELD, FLOAT) — the 7 channels at each position.
|
| - **class_label** (FIELD, STRING) — the sample's class, repeated on every row.
|
|
|
| Converted from the sktime `.ts` format. Nothing is dropped: every dimension,
|
| time point, and label is preserved.
|
|
|
| ## Usage
|
|
|
| Read the `.tsfile` files with the Apache TsFile Java or Python SDK.
|
|
|
| ## Source & license
|
|
|
| - Original dataset: https://huggingface.co/datasets/thuml/Time-Series-Library (subset `SelfRegulationSCP2`)
|
| - Author / publisher: thuml (Tsinghua University)
|
| - Paper: https://arxiv.org/abs/2407.13278
|
| - License: CC BY 4.0
|
| |
