CropRunner dropping rows with duplicate 'gsv_panorama_id'

[PabloMoret]

There's a fault in CropRunner.py, specifically a non-intended behaviour while reading the metadata csv file.

Line 72

df_meta = df_meta.drop_duplicates(subset=['gsv_panorama_id']).to_dict('records')

This code logic is actually dropping labels with the same gsv_panorama_id. Needs to be changed to:

df_meta = df_meta.drop_duplicates(subset=['label_id']).to_dict('records')

Though is an easy-to-fix issue I don't have time to do it now. Just wanted you to be advised.

[misaugstad]

Thank you for pointing that one out!! It was missed because we want to default to using an API request over using a CSV, but we definitely need to fix this backup option!