The way calculations are done internally in floating-point isn't 100% precise.
It follows some kinda IEEE 754 standard.
Also, take a look at its Processing's reference -> http://processing.org/reference/float.html

For displaying any numbers, you can alter them by using number formatting functions like the 1 below:
http://processing.org/reference/nf_.html

Here's your example using it:

final float floatRate = 1.7;
final int intRate = (int) floatRate;

println(intRate);
println(floatRate);
println();

final float floatResult = floatRate - intRate;

println(floatResult);
println( nf(floatResult, 0, 0) );

exit();