Recently someone told me to extract the first two columns of a 2D numpy.ndarray by
firstTwoCols = some2dMatrix[:2]
Where is this notation from and how does it work?
I'm only familiar with the comma separated slicing like
twoCols = some2dMatrix[:,:2]
The : before the comma says to get all rows, and the :2 after the comma says for columns 0 up to but not including 2.
firstTwoCols = some2dMatrix[:2]
This will just extract the first 2 rows with all the columns.
twoCols = some2dMatrix[:,:2] is the one that will extract your first 2 columns for all the rows.
The syntax you describe does not extract the first two columns; it extracts the first two rows. If you specify less slices than the dimension of the array, NumPy treats this as equivalent to all further slices being :, so
arr[:2]
is equivalent to
arr[:2, :]
for a 2D array.
Not sure I understand the question but...
If you do:
>>> Matrix = [[x for x in range(1,5)] for x in range(5)]
>>> Matrix
[[1, 2, 3, 4], [1, 2, 3, 4], [1, 2, 3, 4], [1, 2, 3, 4], [1, 2, 3, 4]]
doing Matrix[:2], it will select the first two list in Matrix, [1, 2, 3, 4], [1, 2, 3, 4]. But if you do:
>>> Matrix[:,:2]
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
TypeError: list indices must be integers, not tuple
But if you work with a Numpy, do:
Matrix = np.array(Matrix)
>>>Matrix[:, :2]
array([[1, 2],
[1, 2],
[1, 2],
[1, 2],
[1, 2]])
