4

I want to generate all binary numbers within a specific range, the code I have so far is:

int2binary = {}
largest_number = pow(2,16)
binary = np.unpackbits(np.array([range(largest_number)],dtype=np.uint16).T,axis=1)

As the numbers generated will be higher than an 8-bit binary number can show, I have changed the dtype from np.uint8 (which works) to np.uint16 which then returns the error:

binary = np.unpackbits(np.array([range(largest_number)],dtype=np.uint16).T,axis=1)
TypeError: Expected an input array of unsigned byte data type

How am I able to fix this error? I have looked on the NumpPy website for their datatypes and uint16 is on there so I am not sure why this isn't working.

Update

Using Abhisek Roy's answer, the error is no longer there. However I forgot to add an import part of the code which is giving a new error:

int2binary[i] = b[i]
IndexError: index 1 is out of bounds for axis 0 with size 1

For the loop:

for i in range(largest_number):
    int2binary[i] = b[i]
Improve this question

3 Answers 3

Reset to default
1

I guess this is what you wanted. Ran this. It works. Check and tell.

import numpy as np
int2binary = {}
largest_number = pow(2,16)
a=(np.array([range(largest_number)],dtype=np.uint8))
b = np.unpackbits(a, axis=1)
for row in b:
    x=row
for i in range(len(x)):
    int2binary[(i)] = x[i]
print(int2binary)
Improve this answer
Sign up to request clarification or add additional context in comments.

5 Comments

Give me a minute.
@morgan Check and tell. If it works, please accept the answer. :)
Yes that does remove the error but calling the in2binary object with the value 255 or 260 returns 1 for 255 and 0 for 260 not the binary representation :(
Let me see. I just cleaned your code. Nothing else.
print("{0:b}".format(255)) Use this if converting to binary is your main issue.
0

A view('uint8') of the uint16 array can do the trick:

#! /usr/bin/python3

import numpy as np

print(np.__version__)
int2binary = {}

largest_number = pow(2,16)

A  = np.array(range(largest_number),dtype=np.uint16)

print(A)
print(A.ndim)
print(A.size)
print(A.flags)
print(A.itemsize)
print(A.nbytes)

B = A.view('uint8')

print(B)
print(B.ndim)
print(B.size)
print(B.flags)
print(B.itemsize)
print(B.nbytes)

binary = np.unpackbits(B)
print(binary)
print(binary.ndim)
print(binary.size)
print(binary.flags)
print(binary.itemsize)
print(binary.nbytes)
Improve this answer

Comments

0 
largest_number = pow(2,binary_dim)
midbinary = []
binary =[]
for i in range(largest_number):
    midbinary.append(format(i, '08b').zfill(16))
for j in range(len(midbinary)):
    b1=[]
    for k in midbinary[j]:
        b1.append(int(k))
    binary.append(b1)
binary = np.array(binary)
for i in range(largest_number):
    int2binary[i] = binary[i]
Improve this answer

1 Comment

just fomat the number and fill zero before it. I dont know why uint16 cannot use np.unpackbits.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.