Implicit conversion from float -20.55194805194806 to int loses precision