soniacq's picture
Update dataset description.
904b6fd verified
metadata
dataset_info:
  features:
    - name: SOURCE_ID
      dtype: int64
    - name: SUB_CODE
      dtype: int64
    - name: FEAT_CODE
      dtype: int64
    - name: STATUS
      dtype: string
    - name: SHAPE_Leng
      dtype: float64
    - name: SHAPE_Area
      dtype: float64
    - name: geometry
      dtype: string
    - name: telecom_dist
      dtype: float64
    - name: telecom_Id
      dtype: float64
    - name: telecom_Reservation Date
      dtype: string
    - name: 'telecom_Franchisee Name '
      dtype: string
    - name: telecom_Status
      dtype: string
    - name: telecom_Installation Date
      dtype: string
    - name: telecom_Pole Class
      dtype: string
    - name: telecom_Borough
      dtype: string
    - name: telecom_X Coord.
      dtype: float64
    - name: telecom_Y Coord.
      dtype: float64
    - name: telecom_Latitude
      dtype: float64
    - name: telecom_Longitude
      dtype: float64
    - name: telecom_Zone
      dtype: string
    - name: telecom_On street
      dtype: string
    - name: telecom_Cross street 1
      dtype: string
    - name: telecom_Cross street 2
      dtype: string
    - name: telecom_Park Advisory
      dtype: string
    - name: telecom_Historic Landmark Advisory
      dtype: string
    - name: telecom_Scenic Landmark Advisory
      dtype: string
    - name: telecom_BID Advisory
      dtype: string
    - name: telecom_School Advisory
      dtype: string
    - name: telecom_Zipcode
      dtype: float64
    - name: telecom_Community Board
      dtype: float64
    - name: telecom_Council District
      dtype: float64
    - name: sign_dist
      dtype: float64
    - name: sign_order_number
      dtype: string
    - name: sign_record_type
      dtype: string
    - name: sign_order_type
      dtype: string
    - name: sign_borough
      dtype: string
    - name: sign_on_street
      dtype: string
    - name: sign_on_street_suffix
      dtype: string
    - name: sign_from_street
      dtype: string
    - name: sign_from_street_suffix
      dtype: string
    - name: sign_to_street
      dtype: string
    - name: sign_to_street_suffix
      dtype: string
    - name: sign_side_of_street
      dtype: string
    - name: sign_order_completed_on_date
      dtype: string
    - name: sign_sign_code
      dtype: string
    - name: sign_sign_description
      dtype: string
    - name: sign_sign_size
      dtype: string
    - name: sign_sign_design_voided_on_date
      dtype: string
    - name: sign_sign_location
      dtype: string
    - name: sign_distance_from_intersection
      dtype: float64
    - name: sign_arrow_direction
      dtype: string
    - name: sign_facing_direction
      dtype: string
    - name: sign_sheeting_type
      dtype: string
    - name: sign_support
      dtype: string
    - name: sign_sign_notes
      dtype: string
    - name: sign_sign_x_coord
      dtype: float64
    - name: sign_sign_y_coord
      dtype: float64
  splits:
    - name: train
      num_bytes: 932386794
      num_examples: 70288
  download_size: 43978813
  dataset_size: 932386794
configs:
  - config_name: default
    data_files:
      - split: train
        path: data/train-*

Fixed Obstructions on Sidewalks (NYC Combined) Dataset

A unified dataset combining NYC’s fixed infrastructure layers relevant to pedestrian and bicycle mobility:

  1. Mobile Telecommunications Franchise Poles — locations of street light poles, traffic light poles, and utility poles reserved for telecommunications installations,
  2. Street Sign Work Orders — locations of regulatory, warning, guide, and parking signs managed by NYC DOT,
  3. NYC Planimetric Sidewalks — official polygon geometries of NYC sidewalks.

These datasets support spatial analysis of pedestrian infrastructure, sidewalk obstruction studies, accessibility assessments, and micro-mobility planning. All records are georeferenced with accurate coordinates or geometries suitable for spatial overlay, proximity analysis, and visualization.

More information about how this dataset was generated can be found here.