README.md

Layers

pz (shapes)

• Manually drawn line segments of varying lengths, each representing a length of curb with particular regulations
• shape: (3349, 18)
• GeoDataFrame (lines)
• column GlobalID is unique identifier

pza (images)

• Related table with links to photos of signs
• shape: (3162, 7)
• DataFrame
• column GLOBALID is unique identifier
• column REL_GLOBALID has 3130 unique values out of 3162 rows (32 duplicate values)
• DATA column has byte image

pzr (rules)

• Related table with a record for each unique regulation (many to one with Parking_Zones lines)
• shape: (4514, 19)
• DataFrame
• column GlobalID is unique identifier
• column GUID has 3203 unique values out of 4514 rows
• GlobalID and GUID never have the same value

Relationships

pz (shapes) to pza (images)

• pz['GlobalID'] joins to pza['REL_GLOBALID'] linking shapes to images
  • 3130 of pz['GlobaID'] (shapes UUID) are in pza['REL_GLOBALID'] (image non-UUID) while 219 aren't
    • i.e. not quite every shape has an image associated with it
  • all 3162 values of pz['REL_GLOBALID'] are in pz['GlobalID']
    • all images have a shape associated with them
    • there are 32 extra images that duplicates for individual street segments)

pz (shapes) to pzr (rules)

• pz['GlobalID'] joins to pzr['GUID']
  • 3203 values of pz['GlobalID'] (shapes) are in pza['GUID'] (rules non-UUID), 146 are not
    • most but not all of the shapes have rules associated with them
    • every unique value of rules has a shape associated with it, every rule has a shape associated with it
  • 4501 values of pzr['GUID'] are in pz['GlobalID'], 13 are not
    • I can't square this one. almost all of the non-unique ids of rules have a shape associates with them. but if the number of unique shapes matches the number of unique values in pzr['GUID'] then how does every rule not have a shape associated with it.

pza (images) to pzr (rules)

• pza['REL_GLOBALID'] joins to pzr['GUID']
  • 4288 values of pzr['GUID'] (non-unique rule identifiers) are in pza['REL_GLOBALID'] (non-unique image identifiers), 226 are not
    • most of the rules have images associated with them but there are more rules than images so there are many cases in which more than one rule is associated with an image
    • there are 226 rules that don't have images associated with them
  • 3030 values of pza['REL_GLOBALID'] (non-unique image identifiers) are in pzr['GUID'] (non-unique rule identifiers), 132 are not
    • most of the images images have rules associated with them. There are 132 images that just don't have any rules associated.

Objectives

• We are trying to create a data set at the street segment (i.e. shapes, pz) level