python - How does slice indexing work in numpy array

Question

Suppose we have an array

a = np.array([[1,2,3,4], [5,6,7,8], [9,10,11,12]])

Now I have below

row_r1 = a[1, :]
row_r2 = a[1:2, :]

print(row_r1.shape)
print(row_r2.shape)

I don't understand why row_r1.shape is (4,) and row_r2.shape is (1,4)

Shouldn't their shape all equal to (4,)?

Answer 1

I like to think of it this way. The first way row[1, :], states go get me all values on row 1 like this:

Returning: array([5, 6, 7, 8])

shape

(4,) Four values in a numpy array.

Where as the second row[1:2, :], states go get me a slice of data between index 1 and index 2:

Returning:

array([[5, 6, 7, 8]]) Note: the double brackets

shape

(1,4) Four values in on one row in a np.array.