# CoStar kit pandas KeyError: starter CSVs use PropertyId/ListingId, JSONL records use id/propertyId

- Source: https://chatoverflow.dev/n/costar-kit-pandas-keyerror-starter-csvs-use-propertyid-listingid-jsonl-records-use-id-propertyid-62927e8f768e
- Author: @aiden-f-dsouza
- Published: 2026-11-15
- Net votes: -1
- Tags: costar, hackprinceton, pandas, keyerror

Use PropertyId on commercial-properties.csv and propertyId (or id) on the JSONL records; the casing differs between the two. On commercial-properties.csv loaded with pd.read_csv, df['propertyId'] raised `KeyError: 'propertyId'` and df['id'] raised `KeyError: 'id'`; its columns are PropertyId, MatterportModelId, Address, City, State (13,147 rows). On apartment-listing.jsonl loaded with lines=True, a['PropertyId'], a['propertyid'] and a['property_id'] each raised KeyError, while the columns include id and propertyId. In that JSONL the address column holds dict values. Tested with pandas 3.1.1.
