How to get the index of the element in the dataframe column that can't be corced to the type of this column?

[mkondratyev85]

I'm trying to use pandera for validate dataframes that sometimes have wrong values in certain cells.
I would like to print out the exact positions of the cells that were casing the error. But it seems like I can't get this info from the errors that I catch with try-except block.

Here is the example code:

import pandas as pd
import pandera as pa

df = pd.DataFrame({
    "X": [1, 2, 3],
    "Y": [1, 'a', 3],
    "Z": [1, 2, 3],
})


# define schema
schema = pa.DataFrameSchema({
    "X": pa.Column(float),
    "Y": pa.Column(float),
    "Z": pa.Column(float),
    },
    index=pa.Index(str),
    coerce=True,
    )

try:
    sections = schema(df, lazy=True)
except pa.errors.SchemaErrors as err:
    print(err.failure_cases)
    print(err.data)

And here is the output:

  schema_context column                    check check_number failure_case index
0         Column      Y  coerce_dtype('float64')         None       object  None
1         Column      Y             not_nullable         None         None     0
2         Column      Y             not_nullable         None         None     1
3         Column      Y             not_nullable         None         None     2
     X     Y    Z
0  1.0  None  1.0
1  2.0  None  2.0
2  3.0  None  3.0

As you can see, there is no information about the index of the problem element in column Y.
Is there a way to get this position?

And why are all other elements in Y column is also have the problem now (not_nullable) while they can be represented as float?

EDIT:
After researching this issue for a while, I figured out that the problem is on the stage of trying to coerce values in column Y to the float type. If I change the invalid dataframe to be with all number values but add additional check for the field so that some values in my invalid dateframe would not pass, then, I would get the index of the position of the wrong element in the dataframe:

df = pd.DataFrame({
    "X": [1, 2, 3],
    "Y": [1, -10, 3],
    "Z": [1, 2, 3],
})

schema = pa.DataFrameSchema({
    "X": pa.Column(float, checks=[pa.Check(lambda x: x > 0, element_wise=True)]),
    "Y": pa.Column(float, checks=[pa.Check(lambda x: x > 0, element_wise=True)]),
    "Z": pa.Column(float, checks=[pa.Check(lambda x: x > 0, element_wise=True)]),
    },
    index=pa.Index(str),
    coerce=True,
    )

So my question could be narrowed down a bit like this:

How to get the index of the element in the dataframe column that can't be corced to the type of this column?