[REFACTOR]: Data validation of properties.

[arv-anshul]

As you know the fetched data is not well type hinted that's why the streamlit app throw ValidationError which is annoying. So remove the pydantic validation on the fetched data.

😎 A better approach!

# Whole fetched data converted as pd.DataFrame
df = fetched_data_as_dataframe

imp_cols = [
    'PROP_ID', 'PREFERENCE', 'DESCRIPTION', 'PROPERTY_TYPE', 'CITY', 'TRANSACT_TYPE',
    'OWNTYPE', 'CLASS_LABEL', 'BEDROOM_NUM', 'PRICE_PER_UNIT_AREA', 'FURNISH',
    'FACING', 'AGE', 'TOTAL_FLOOR', 'FEATURES', 'PROP_NAME', 'PRICE_SQFT', 'MAP_DETAILS',
    'AMENITIES', 'AREA', 'PRICE', 'PROP_HEADING', 'SECONDARY_TAGS', 'TOTAL_LANDMARK_COUNT',
    'FORMATTED_LANDMARK_DETAILS', 'SOCIETY_NAME', 'BUILDING_NAME', 'location', 'BALCONY_NUM',
    'FLOOR_NUM', 'CARPET_SQFT', 'SUPERBUILTUP_SQFT', 'BUILTUP_SQFT', 'SUPER_SQFT',
]

# Filter the data using the most important columns
df = df[imp_cols]  # Now df contains only important columns

Now, you can provide a multiselect option in the app from that users can choose other important data (column/feature) related to properties. As specified above this method remove the pydantic validation on the fetched data.