double to float without rounding python

Question

say a = 580991.3636

When I b = np.float32(a)

b 580991.4

I'm after b 580991.3636

I'm hoping this is an easy one, any ideas? I'm also assuming it is my misunderstanding of floats causing this. I know python is dynamic when it comes to dtypes, but I need it to be in float32 for a specific library (python-pcl).

The closest float to 580991.3636 has exact value 580991.375 which would print as 580991.4 — Patricia Shanahan
– Patricia Shanahan, Commented Nov 25, 2018 at 16:47

user9849588 · Accepted Answer · 2018-11-25 16:50:01Z

3

Float32 can only offer between 6 and 9 digits of precision.

Float32 works by storing the sign in 1 bit, the exponent in 8 bits and the value in 23 bits (also called the fraction). So you can represent a wide range of real numbers at the cost of precision.

If you need it to be more precise it has to be a double or use the python's library decimal.

edited Nov 25, 2018 at 16:50

answered Nov 25, 2018 at 16:43

user9849588

6071 gold badge7 silver badges13 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

double to float without rounding python

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related