The behaviour of int(2.8) returning 2 shows that int() on a float:

• arounds to the nearest integer
• bdiscards the fractional part (truncates toward zero)
• calways rounds up
• draises an error