What does the syntax ndarray[:,list[i]] means in python?

Question

I was looking at a tutorial on the web for Python. I don't know anything about Python, so I searched and could not find the answer.

There is some code like this:

s = np.tanh(self.X[:,Y[t]])

Where, X is ndarray and Y is a List of lists (where each list is integer-type), np is a numpy object, and tanh is the hyperbolic tangent.

What does this syntax mean?

X is almost certainly not a list of lists, but probably a numpy ndarray. — DSM
– DSM, Commented Apr 2, 2016 at 12:03
@DSM X was generated using "np.random.uniform". Sorry, as I said, I am new to python. — Hemendra Sharma
– Hemendra Sharma, Commented Apr 2, 2016 at 12:08

bakkal · Accepted Answer · 2016-04-02 13:01:36Z

2

In the context of numpy, it can e.g. allow to access the columns, so e.g. in your example X[:, Y[t]], it allows you to access the column of X, indexed by the value in Y[t].

The : basically says "all rows" and the Y[t] specifies the column index.

Here's a simple example to see it in action:

In [1]: import numpy as np

In [3]: m = np.array([['a', 'b'], ['c', 'd'], ['f', 'g']])

In [4]: m[:, 0]
Out[4]: 
array(['a', 'c', 'f'], 
      dtype='|S1')

In [5]: m[:, 1]
Out[5]: 
array(['b', 'd', 'g'], 
      dtype='|S1')

What if the "column index" is a list?

If you use m[:, some_list], the : colon would ask for all rows, and then the columns would be the column indices in some_list, in that order

so e.g. if we want all rows, and the columns [1, 0] (in that order), here's what you get:

In [53]: m[:, [1, 0]]
Out[53]: 
array([['b', 'a'],
       ['d', 'c'],
       ['g', 'f']], 
      dtype='|S1')

edited Apr 2, 2016 at 13:01

answered Apr 2, 2016 at 12:01

bakkal

55.6k12 gold badges136 silver badges113 bronze badges

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

2 Comments

Hemendra Sharma Over a year ago

Actually Y[t] is also a list. That means, Y itself is a list-of-lists. So, Y[t] would be [2,4,5,6,8,...]. What would be the output of X[:,Y[t]] is that case? Possible guess ! The output can be... m[ : , [0,1] ] => [ 'a' , 'c' , 'f' , 'b' , 'd' , 'g' ] Am I right?

bakkal Over a year ago

@HemendraSharma m[:, [0,1]] would be asking for "all rows of m, in column 0 and column 1", I added the trace in my answer to explain

Collectives™ on Stack Overflow

What does the syntax ndarray[:,list[i]] means in python?

1 Answer 1

What if the "column index" is a list?

2 Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

What if the "column index" is a list?

2 Comments

Your Answer

Sign up or log in

Post as a guest

Related