Datasets:

THUgewu commited on
Commit
8ee20ca
·
verified ·
1 Parent(s): 42dab19

chore: sync metadata, drop original files

Browse files
Files changed (1) hide show
  1. README.md +70 -65
README.md CHANGED
@@ -1,65 +1,70 @@
1
- ---
2
- license: cc-by-4.0
3
- task_categories:
4
- - time-series-forecasting
5
- tags:
6
- - classification
7
- - forecasting
8
- - anomaly-detection
9
- - TSLib
10
- - tsfile
11
- - modality:timeseries
12
- pretty_name: SelfRegulationSCP1 (TsFile)
13
- size_categories:
14
- - 1M<n<10M
15
- configs:
16
- - config_name: default
17
- data_files:
18
- - split: train
19
- path: SelfRegulationSCP1_train.tsfile
20
- - split: test
21
- path: SelfRegulationSCP1_test.tsfile
22
- modality:
23
- - timeseries
24
- language:
25
- - en
26
- ---
27
-
28
- # SelfRegulationSCP1 (TsFile)
29
-
30
- Apache TsFile version of the `SelfRegulationSCP1` UEA classification subset of
31
- [`thuml/Time-Series-Library`](https://huggingface.co/datasets/thuml/Time-Series-Library).
32
-
33
- ## Overview
34
-
35
- EEG slow cortical potentials (healthy subject); classify cursor up vs down.
36
-
37
- - **Train samples:** 268 • **Test samples:** 293.
38
- - **Dimensions (channels):** 6 • **Series length:** 896.
39
- - **Classes:** 2.
40
-
41
- Each sample is an independent multivariate series. TRAIN and TEST are stored as
42
- two separate TsFiles (`SelfRegulationSCP1_train.tsfile` / `SelfRegulationSCP1_test.tsfile`).
43
-
44
- ## Schema (TsFile structure)
45
-
46
- - **sample_index** (TAG) — one device per sample. Query one sample with
47
- `WHERE sample_index=0`.
48
- - **Time** (INT64) — within-sample position (0..895); a frame index, not a
49
- wall-clock timestamp (the sktime `.ts` source has none).
50
- - **dim_0 … dim_5** (FIELD, FLOAT) — the 6 channels at each position.
51
- - **class_label** (FIELD, STRING) — the sample's class, repeated on every row.
52
-
53
- Converted from the sktime `.ts` format. Nothing is dropped: every dimension,
54
- time point, and label is preserved.
55
-
56
- ## Usage
57
-
58
- Read the `.tsfile` files with the Apache TsFile Java or Python SDK.
59
-
60
- ## Source & license
61
-
62
- - Original dataset: https://huggingface.co/datasets/thuml/Time-Series-Library (subset `SelfRegulationSCP1`)
63
- - Author / publisher: thuml (Tsinghua University)
64
- - Paper: https://arxiv.org/abs/2407.13278
65
- - License: CC BY 4.0
 
 
 
 
 
 
1
+ ---
2
+ license: cc-by-4.0
3
+ task_categories:
4
+ - time-series-forecasting
5
+ tags:
6
+ - time-series
7
+ - forecasting
8
+ - anomaly-detection
9
+ - classification
10
+ - TSLib
11
+ - tsfile
12
+ - modality:timeseries
13
+ - timeseries
14
+ - format:tsfile
15
+ pretty_name: SelfRegulationSCP1 (TsFile)
16
+ size_categories:
17
+ - 1M<n<10M
18
+ configs:
19
+ - config_name: default
20
+ data_files:
21
+ - split: train
22
+ path: SelfRegulationSCP1_train.tsfile
23
+ - split: test
24
+ path: SelfRegulationSCP1_test.tsfile
25
+ modality:
26
+ - timeseries
27
+ - tabular
28
+ - text
29
+ language:
30
+ - en
31
+ ---
32
+
33
+ # SelfRegulationSCP1 (TsFile)
34
+
35
+ Apache TsFile version of the `SelfRegulationSCP1` UEA classification subset of
36
+ [`thuml/Time-Series-Library`](https://huggingface.co/datasets/thuml/Time-Series-Library).
37
+
38
+ ## Overview
39
+
40
+ EEG slow cortical potentials (healthy subject); classify cursor up vs down.
41
+
42
+ - **Train samples:** 268 • **Test samples:** 293.
43
+ - **Dimensions (channels):** 6 • **Series length:** 896.
44
+ - **Classes:** 2.
45
+
46
+ Each sample is an independent multivariate series. TRAIN and TEST are stored as
47
+ two separate TsFiles (`SelfRegulationSCP1_train.tsfile` / `SelfRegulationSCP1_test.tsfile`).
48
+
49
+ ## Schema (TsFile structure)
50
+
51
+ - **sample_index** (TAG) — one device per sample. Query one sample with
52
+ `WHERE sample_index=0`.
53
+ - **Time** (INT64) — within-sample position (0..895); a frame index, not a
54
+ wall-clock timestamp (the sktime `.ts` source has none).
55
+ - **dim_0 … dim_5** (FIELD, FLOAT) — the 6 channels at each position.
56
+ - **class_label** (FIELD, STRING) — the sample's class, repeated on every row.
57
+
58
+ Converted from the sktime `.ts` format. Nothing is dropped: every dimension,
59
+ time point, and label is preserved.
60
+
61
+ ## Usage
62
+
63
+ Read the `.tsfile` files with the Apache TsFile Java or Python SDK.
64
+
65
+ ## Source & license
66
+
67
+ - Original dataset: https://huggingface.co/datasets/thuml/Time-Series-Library (subset `SelfRegulationSCP1`)
68
+ - Author / publisher: thuml (Tsinghua University)
69
+ - Paper: https://arxiv.org/abs/2407.13278
70
+ - License: CC BY 4.0