Pandas or Polars dataframes assert for equality with tolerance on float precision?

[morindo]

Is there a way to assert two Pandas or Polars dataframes for equality and have tolerance on float precision?

Right now when I save a Pandas dataframe as a snaphot using df.to_dict(), I'm able to compare actual and snapshot dataframes for equality on my Windows PC with success, but when I run pytest on a Linux machine, float columns are different and it fail the comparison due to float precision.

[noahnu]

I believe this is a duplicate of #913. Let me know if I've misinterpreted.

[morindo]

I confirm, the propose solution in syrupy v5 is working great!

def test_round_float(snapshot):
    assert (2.0 / 3.0) == snapshot(
        matcher=path_type(
            types=(float,), replacer=lambda data, _: round(data, 5)
        )
    )