---
license: apache-2.0
dataset_info:
  features:
  - name: id
    dtype: int32
  - name: x
    dtype:
      array2_d:
        shape:
        - 24
        - 13
        dtype: float32
  - name: station
    dtype:
      class_label:
        names:
          '0': Aotizhongxin
          '1': Changping
          '2': Dingling
          '3': Dongsi
          '4': Guanyuan
          '5': Gucheng
          '6': Huairou
          '7': Nongzhanguan
          '8': Shunyi
          '9': Tiantan
          '10': Wanliu
          '11': Wanshouxigong
  - name: year
    dtype:
      class_label:
        names:
          '0': '2013'
          '1': '2014'
          '2': '2015'
          '3': '2016'
          '4': '2017'
  - name: month
    dtype:
      class_label:
        names:
          '0': '1'
          '1': '2'
          '2': '3'
          '3': '4'
          '4': '5'
          '5': '6'
          '6': '7'
          '7': '8'
          '8': '9'
          '9': '10'
          '10': '11'
          '11': '12'
  - name: day
    dtype:
      class_label:
        names:
          '0': '1'
          '1': '2'
          '2': '3'
          '3': '4'
          '4': '5'
          '5': '6'
          '6': '7'
          '7': '8'
          '8': '9'
          '9': '10'
          '10': '11'
          '11': '12'
          '12': '13'
          '13': '14'
          '14': '15'
          '15': '16'
          '16': '17'
          '17': '18'
          '18': '19'
          '19': '20'
          '20': '21'
          '21': '22'
          '22': '23'
          '23': '24'
          '24': '25'
          '25': '26'
          '26': '27'
          '27': '28'
          '28': '29'
          '29': '30'
          '30': '31'
  - name: season
    dtype:
      class_label:
        names:
          '0': Summer
          '1': Autumn
          '2': Winter
          '3': Spring
  splits:
  - name: train
    num_bytes: 17082624
    num_examples: 12272
  - name: val
    num_bytes: 3660960
    num_examples: 2630
  - name: test
    num_bytes: 3660960
    num_examples: 2630
  download_size: 9437882
  dataset_size: 24404544
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: val
    path: data/val-*
  - split: test
    path: data/test-*
---

## MSD Beijing Multi-Site Air Quality Dataset Attribution

The Multi-factor Sequential Disentanglement benchmark includes the **Beijing Multi-Site Air Quality (BMS-AQ) dataset**, a time series dataset that captures daily air quality and weather measurements across multiple monitoring stations in Beijing.
For the benchmark, we preprocess this data into daily sequences of 24 hourly records, grouped by station and date.
Each sequence is labeled with static attributes such as station, year, month, day, and season, resulting in a temporal dataset suitable for studying disentanglement in real-world time series.

- Original source:  
  https://archive.ics.uci.edu/dataset/501/beijing+multi+site+air+quality+data

  ```
  @misc{beijing_multi-site_air_quality_501,
    author       = {Chen, Song},
    title        = {{Beijing Multi-Site Air Quality}},
    year         = {2017},
    howpublished = {UCI Machine Learning Repository},
    note         = {{DOI}: https://doi.org/10.24432/C5RK5G}
  }
  ```
  

⚠ **Note:** We process and redistribute this dataset solely for non-commercial research purposes. Please cite the above paper when using this dataset in your work.