How to read a binary file of type complex64 values in Python

Question

I have a binary file that contains several complex numbers of type complex64? (i.e. four bytes of type float for the real part and another four bytes for the imaginary part). The real and imaginary parts are multiplexed so that the real part is stored first and followed by the imaginary part.

complex64 is a complex number represented by two 32-bit floats (real and imaginary components). Is it what you have in your file? — Jacques Gaudin
– Jacques Gaudin, Commented Jun 26, 2016 at 23:58
@Moses, I don't know if there is a way to upload here, but the complex values in the file have been written from a program called GNURadio, and the docs say that a binary file of complex numbers stores the real and imag parts as 32-bits floats (first element is the real and the second is the imag) I tried to read the file using the following command but I could not get the numbers that I was expecting: numpy.fromfile('file_name', dtype=complex) @Jacques, yes this is what I have. I was expecting [0, 2+j, -3.14-7.99j] but I get something like: [0 +7.81e-3j, -1.29e+5 +0j, 7.8e-3 -1.29e+5j] — Doe
– Doe, Commented Jun 27, 2016 at 6:09
The doc is here:nutaq.com/blog/gnu-radio-file-source-and-sink. Why don't you do numpy.fromfile('file_name', dtype=numpy.complex64)? — Jacques Gaudin
– Jacques Gaudin, Commented Jun 27, 2016 at 9:12
gnuradio.org/redmine/projects/gnuradio/wiki/… seems to indicate the same use as my comment above — Jacques Gaudin
– Jacques Gaudin, Commented Jun 27, 2016 at 11:24

Jacques Gaudin · Accepted Answer · 2016-06-28 17:40:45Z

2

I was able to reproduce the error you encounter by creating an array of complex64 from [0, 2+j, -3.14-7.99j], saving it to a file and reading it as Python built-in complex type.

The issue is that the built-in complex type has the size of a C double which, depending on your plateform, may differ from 32-bits (256 bits on my machine).

You must use numpy.fromfile('file_name', dtype=numpy.complex64) to read your file correctly, i.e. make sure the complex numbers are read as two 32-bits floating point numbers.

edited Jun 28, 2016 at 17:40

answered Jun 28, 2016 at 16:53

Jacques Gaudin

17.1k11 gold badges59 silver badges80 bronze badges

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

Collectives™ on Stack Overflow

How to read a binary file of type complex64 values in 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