Datasetpublic
thuml/Time-Series-Library

Time-Series-Library (TSLib) TSLib is an open-source library for deep learning researchers, especially for deep time series analysis. We provide a neat code base to evaluate advanced deep time series models or develop your model, which covers five mainstream tasks: long- and short-term forecasting, imputation, anomaly detection, and classification. This benchmark collection is designed to evaluate and develop advanced deep time-series models. For an in-depth exploration of… See the full description on the dataset page: https://huggingface.co/datasets/thuml/Time-Series-Library.
sourceHugging Facecc-by-4.0updated 11mo agoView on Hugging Face
9likes23kdownloads
rootREADME.md
README.md312 linesDownload Raw Back to root
1---2tags:3- time-series4- forecasting5- anomaly-detection6- classification7- TSLib8license: cc-by-4.09task_categories:10- time-series-forecasting11pretty_name: Time-Series-Library (TSLib)12language:13- en14configs:15- config_name: ETTh116  description: ETT long-term forecasting subset ETTh1 (hourly).17  data_files:18  - ETT-small/ETTh1.csv19- config_name: ETTh220  description: ETT long-term forecasting subset ETTh2 (hourly).21  data_files:22  - ETT-small/ETTh2.csv23- config_name: ETTm124  description: ETT long-term forecasting subset ETTm1 (15-min).25  data_files:26  - ETT-small/ETTm1.csv27- config_name: ETTm228  description: ETT long-term forecasting subset ETTm2 (15-min).29  data_files:30  - ETT-small/ETTm2.csv31- config_name: electricity32  description: Electricity load forecasting (UCI Electricity).33  data_files:34  - electricity/electricity.csv35- config_name: traffic36  description: Traffic volume forecasting.37  data_files:38  - traffic/traffic.csv39- config_name: weather40  description: Weather time-series forecasting.41  data_files:42  - weather/weather.csv43- config_name: exchange_rate44  description: Exchange rate forecasting.45  data_files:46  - exchange_rate/exchange_rate.csv47- config_name: national_illness48  description: Influenza-like illness (ILI) forecasting.49  data_files:50  - illness/national_illness.csv51- config_name: m4-yearly52  description: M4 Yearly forecasting subset.53  data_files:54  - split: train55    path: m4/Yearly-train.csv56  - split: test57    path: m4/Yearly-test.csv58- config_name: m4-quarterly59  description: M4 Quarterly forecasting subset.60  data_files:61  - split: train62    path: m4/Quarterly-train.csv63  - split: test64    path: m4/Quarterly-test.csv65- config_name: m4-monthly66  description: M4 Monthly forecasting subset.67  data_files:68  - split: train69    path: m4/Monthly-train.csv70  - split: test71    path: m4/Monthly-test.csv72- config_name: m4-weekly73  description: M4 Weekly forecasting subset.74  data_files:75  - split: train76    path: m4/Weekly-train.csv77  - split: test78    path: m4/Weekly-test.csv79- config_name: m4-daily80  description: M4 Daily forecasting subset.81  data_files:82  - split: train83    path: m4/Daily-train.csv84  - split: test85    path: m4/Daily-test.csv86- config_name: m4-hourly87  description: M4 Hourly forecasting subset.88  data_files:89  - split: train90    path: m4/Hourly-train.csv91  - split: test92    path: m4/Hourly-test.csv93- config_name: EthanolConcentration94  description: 'UEA multivariate classification: EthanolConcentration.'95  data_files:96  - split: train97    path: EthanolConcentration/EthanolConcentration_TRAIN.ts98  - split: test99    path: EthanolConcentration/EthanolConcentration_TEST.ts100- config_name: FaceDetection101  description: 'UEA multivariate classification: FaceDetection.'102  data_files:103  - split: train104    path: FaceDetection/FaceDetection_TRAIN.ts105  - split: test106    path: FaceDetection/FaceDetection_TEST.ts107- config_name: Handwriting108  description: 'UEA multivariate classification: Handwriting.'109  data_files:110  - split: train111    path: Handwriting/Handwriting_TRAIN.ts112  - split: test113    path: Handwriting/Handwriting_TEST.ts114- config_name: Heartbeat115  description: 'UEA multivariate classification: Heartbeat.'116  data_files:117  - split: train118    path: Heartbeat/Heartbeat_TRAIN.ts119  - split: test120    path: Heartbeat/Heartbeat_TEST.ts121- config_name: JapaneseVowels122  description: 'UEA multivariate classification: JapaneseVowels.'123  data_files:124  - split: train125    path: JapaneseVowels/JapaneseVowels_TRAIN.ts126  - split: test127    path: JapaneseVowels/JapaneseVowels_TEST.ts128- config_name: PEMS-SF129  description: 'UEA multivariate classification: PEMS-SF.'130  data_files:131  - split: train132    path: PEMS-SF/PEMS-SF_TRAIN.ts133  - split: test134    path: PEMS-SF/PEMS-SF_TEST.ts135- config_name: SelfRegulationSCP1136  description: 'UEA multivariate classification: SelfRegulationSCP1.'137  data_files:138  - split: train139    path: SelfRegulationSCP1/SelfRegulationSCP1_TRAIN.ts140  - split: test141    path: SelfRegulationSCP1/SelfRegulationSCP1_TEST.ts142- config_name: SelfRegulationSCP2143  description: 'UEA multivariate classification: SelfRegulationSCP2.'144  data_files:145  - split: train146    path: SelfRegulationSCP2/SelfRegulationSCP2_TRAIN.ts147  - split: test148    path: SelfRegulationSCP2/SelfRegulationSCP2_TEST.ts149- config_name: SpokenArabicDigits150  description: 'UEA multivariate classification: SpokenArabicDigits.'151  data_files:152  - split: train153    path: SpokenArabicDigits/SpokenArabicDigits_TRAIN.ts154  - split: test155    path: SpokenArabicDigits/SpokenArabicDigits_TEST.ts156- config_name: UWaveGestureLibrary157  description: 'UEA multivariate classification: UWaveGestureLibrary.'158  data_files:159  - split: train160    path: UWaveGestureLibrary/UWaveGestureLibrary_TRAIN.ts161  - split: test162    path: UWaveGestureLibrary/UWaveGestureLibrary_TEST.ts163- config_name: SMD-data164  description: Server Machine Dataset (SMD) for anomaly detection — train & test data.165  data_files:166  - split: train167    path: SMD/SMD_train.npy168  - split: test169    path: SMD/SMD_test.npy170- config_name: SMD-label171  description: Server Machine Dataset (SMD) — test anomaly labels.172  data_files:173  - split: test_label174    path: SMD/SMD_test_label.npy175- config_name: MSL-data176  description: NASA Mars Science Laboratory (MSL) anomaly detection — train/test arrays.177  data_files:178  - split: train179    path: MSL/MSL_train.npy180  - split: test181    path: MSL/MSL_test.npy182- config_name: MSL-label183  description: MSL anomaly detection — test labels.184  data_files:185  - split: test_label186    path: MSL/MSL_test_label.npy187- config_name: SMAP-data188  description: >-189    NASA Soil Moisture Active Passive (SMAP) anomaly detection — train/test190    arrays.191  data_files:192  - split: train193    path: SMAP/SMAP_train.npy194  - split: test195    path: SMAP/SMAP_test.npy196- config_name: SMAP-label197  description: SMAP anomaly detection — test labels.198  data_files:199  - split: test_label200    path: SMAP/SMAP_test_label.npy201- config_name: PSM-data202  description: KPI-based Process/System Monitoring data (train/test).203  data_files:204  - split: train205    path: PSM/train.csv206  - split: test207    path: PSM/test.csv208- config_name: PSM-label209  description: KPI-based Process/System Monitoring labels (test_label).210  data_files:211  - split: test_label212    path: PSM/test_label.csv213- config_name: SWaT214  description: Secure Water Treatment (SWaT) anomaly detection, processed data.215  data_files:216  - split: train217    path: SWaT/swat_train2.csv218  - split: test219    path: SWaT/swat2.csv220size_categories:221- 10M<n<100M222---223 224 225# Time-Series-Library (TSLib)226 227TSLib is an open-source library for deep learning researchers, especially for deep time series analysis.228 229We provide a neat code base to evaluate advanced deep time series models or develop your model, which covers five mainstream tasks: **long- and short-term forecasting, imputation, anomaly detection, and classification.**230 231This benchmark collection is designed to evaluate and develop advanced deep time-series models. For an in-depth exploration of current time-series models and their performance, please refer to our paper **[Deep Time Series Models: A Comprehensive Survey and Benchmark](https://arxiv.org/abs/2407.13278)**.232 233To get started with the codebase and contribute, please visit the **[GitHub repository](https://github.com/thuml/Time-Series-Library)**.234 235## Dataset Overview236 237| **Tasks**         | **Benchmarks**                                                                 | **Metrics**                          | **Series Length**     |238|-------------------|-------------------------------------------------------------------------------|--------------------------------------|-----------------------|239| **Forecasting**    | **Long-term:** ETT (4 subsets), Electricity, Traffic, Weather, Exchange, ILI  | MSE, MAE                             | 96\~720 (ILI: 24\~60)   |240|                   | **Short-term:** M4 (6 subsets)                                                | SMAPE, MASE, OWA                    | 6\~48                  |241| **Imputation**     | ETT (4 subsets), Electricity, Weather                                        | MSE, MAE                             | 96                    |242| **Classification** | UEA (10 subsets)                                                              | Accuracy                             | 29\~1751               |243| **Anomaly Detection** | SMD, MSL, SMAP, SWaT, PSM                                                  | Precision, Recall, F1-Score         | 100                   |244 245 246## File Structure247```248Time-Series-Library/249├── ETT-small/250├── EthanolConcentration/251├── FaceDetection/252├── Handwriting/253├── Heartbeat/254├── JapaneseVowels/255├── MSL/256├── PEMS-SF/257├── PSM/258├── SMAP/259├── SMD/260├── SWaT/261├── SelfRegulationSCP1/262├── SelfRegulationSCP2/263├── SpokenArabicDigits/264├── UWaveGestureLibrary/265├── electricity/266├── exchange_rate/267├── illness/268├── m4/269├── traffic/270├── weather/271├── .gitattributes272└── README.md273```274 275## Usage276 277You can load the dataset directly using the `datasets` library:278 279```280from datasets import load_dataset281dataset = load_dataset("thuml/Time-Series-Library", "ETTh1")282```283 284Or download specific files with hf_hub_download:285 286```287from huggingface_hub import hf_hub_download288hf_hub_download("thuml/Time-Series-Library", "ETT-small/ETTh1.csv", repo_type="dataset")289```290 291## License292This dataset is released under the CC BY 4.0 License.293 294## Citation295 296If you find this repo useful, please cite our paper.297 298```299@inproceedings{wu2023timesnet,300  title={TimesNet: Temporal 2D-Variation Modeling for General Time Series Analysis},301  author={Haixu Wu and Tengge Hu and Yong Liu and Hang Zhou and Jianmin Wang and Mingsheng Long},302  booktitle={International Conference on Learning Representations},303  year={2023},304}305 306@article{wang2024tssurvey,307  title={Deep Time Series Models: A Comprehensive Survey and Benchmark},308  author={Yuxuan Wang and Haixu Wu and Jiaxiang Dong and Yong Liu and Mingsheng Long and Jianmin Wang},309  booktitle={arXiv preprint arXiv:2407.13278},310  year={2024},311}312```