Convert numpy array with floats to binary (0 or 1 integers)

Question

I have an input array that looks like this:

  [[  0.    1. ]
   [ 10.    0.4]
   [ 20.    1.4]
   [ 30.    3. ]
   [ 40.    1.1]
   [ 50.    0.7]]

Now I'd like to convert the float values from the second column (the ones in array[:, 1]) to single bit binary values represented as 1 or 0 integers. I have threshold value that I'd like to use as a limit between logical 0 and logical 1. Let's say it is 1.5. After the conversion the array should look like this:

  [[  0.    0 ]
   [ 10.    0]
   [ 20.    0]
   [ 30.    1]
   [ 40.    0]
   [ 50.    0]]

How do I do that with the least effort?

To put actual integer values in the array (as opposed to float 1.0) you have switch to a structured array, or just put the binary values in a separate array. — hpaulj
– hpaulj, Commented Jan 16, 2017 at 17:27

Divakar · Accepted Answer · 2017-01-16 08:42:03Z

5

Compare the second column against the threshold, which would be a boolean array and then assign it back to the second column. The assignment would upcast it to float data before assigning back. The resultant second column would still be float, but as 0s and 1s as it needs to maintain the datatype there.

Thus, simply do -

a[:,1] = a[:,1]>1.5

Sample run -

In [47]: a
Out[47]: 
array([[  0. ,   1. ],
       [ 10. ,   0.4],
       [ 20. ,   1.4],
       [ 30. ,   3. ],
       [ 40. ,   1.1],
       [ 50. ,   0.7]])

In [48]: a[:,1] = a[:,1]>1.5

In [49]: a
Out[49]: 
array([[  0.,   0.],
       [ 10.,   0.],
       [ 20.,   0.],
       [ 30.,   1.],
       [ 40.,   0.],
       [ 50.,   0.]])

edited Jan 16, 2017 at 8:42

answered Jan 16, 2017 at 8:38

Divakar

222k19 gold badges273 silver badges374 bronze badges

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

3 Comments

NoDataDumpNoContribution Over a year ago

To be even more precise: The result of a[:,1]>1.5 is a boolean array, but the assignment converts it back to float.

Divakar Over a year ago

@Trilarion Right. Added that.

ThoWe Over a year ago

Yes, it's a boolean array. For instance a = a[:, 1]>1.5 results in [False False False True False False]. But as I need to do math with these values I need them to be integers. But a[:, 1] = a[:, 1]>1.5 works.

Collectives™ on Stack Overflow

Convert numpy array with floats to binary (0 or 1 integers)

1 Answer 1

3 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

3 Comments

Your Answer

Sign up or log in

Post as a guest

Related