Example for docs for dropping Invalid Rows does not seem obvious

[ttimbers]

I have tried the example in the docs here for dropping invalid rows using DataFrameSchema - which leads one to believe that two rows should be dropped (all those less than 3), but the output is the entire data frame. I see after some time, the issue is due to the quotes around the numbers in the construction of the data frame, which makes the column the wrong type (object, not int). I think that is not the intention here... Should those quotes be removed? There is a similar issue I think for SeriesSchema. The example given for dropping invalid rows with DataFrameModels doesn't have this issue and seems to work as obviously intended.

Problematic example

import pandas as pd
import pandera as pa

from pandera import Check, Column, DataFrameSchema

df = pd.DataFrame({"counter": ["1", "2", "3"]})
schema = DataFrameSchema(
    {"counter": Column(int, checks=[Check(lambda x: x >= 3)])},
    drop_invalid_rows=True,
)

schema.validate(df, lazy=True)

output:

  counter
0       1
1       2
2       3

Maybe change to:

import pandas as pd
import pandera as pa

from pandera import Check, Column, DataFrameSchema

df = pd.DataFrame({"counter": [1, 2, 3]})
schema = DataFrameSchema(
    {"counter": Column(int, checks=[Check(lambda x: x >= 3)])},
    drop_invalid_rows=True
)

schema.validate(df, lazy=True)

output:

   counter
2        3