Converting string of "bytes" to numpy array of uint8

Question

I'm using a API that returns RGB data as strings(*) (for instance I get 'ABC' for [65, 66, 67]. Is there a way to have this directly converted to a numpy unint8 array without an explicit comprehension with ord()? Since this is picture data, I could be processing several million bytes, so any shortcut can help.

Otherwise, any method faster than the comprehension is gladly accepted.

(*) API requires Python 2.7, for the time being

can you give a little more of a minimal reproducible example? Like what's the shape of the data array? — Daniel F
– Daniel F, Commented Dec 4, 2017 at 12:12

Paul Panzer · Accepted Answer · 2017-12-04 12:34:57Z

7

You can use np.frombuffer:

np.frombuffer(b'ABC', dtype=np.uint8)
# array([65, 66, 67], dtype=uint8)

Since you are on Python2 this will probably work directly on strings, on Python3 a string would have to be encoded first to yield a bytes object.

edited Dec 4, 2017 at 12:34

answered Dec 4, 2017 at 12:26

Paul Panzer

53.3k3 gold badges59 silver badges103 bronze badges

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

1 Comment

xenoid Over a year ago

Exactly what the doctor ordered... And yes, in Python2 it works directly on strings (I assume that in Python3, the API will give bytes and not strings). In addition, I figured out that I could use reshape() to get an array of RGB triplets.

Collectives™ on Stack Overflow

Converting string of "bytes" to numpy array of uint8

1 Answer 1

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related